diff --git a/..gitignore.un~ b/..gitignore.un~ deleted file mode 100644 index c7ffe5a..0000000 Binary files a/..gitignore.un~ and /dev/null differ diff --git a/.gitignore~ b/.gitignore~ deleted file mode 100644 index 19e30d6..0000000 --- a/.gitignore~ +++ /dev/null @@ -1,5 +0,0 @@ -.ipynb_checkpoints/* -.ipynb_checkpoints -df_realestate.csv -df_realestate_processed.csv -__pycache__ diff --git a/02_PreProcessing/main.ipynb b/02_PreProcessing/main02.ipynb similarity index 100% rename from 02_PreProcessing/main.ipynb rename to 02_PreProcessing/main02.ipynb diff --git a/03_ClusteringTheory/main.ipynb b/03_ClusteringTheory/main03.ipynb similarity index 100% rename from 03_ClusteringTheory/main.ipynb rename to 03_ClusteringTheory/main03.ipynb diff --git a/04_ClusteringPracticeWord2Vec/main.ipynb b/04_ClusteringPracticeWord2Vec/main04.ipynb similarity index 100% rename from 04_ClusteringPracticeWord2Vec/main.ipynb rename to 04_ClusteringPracticeWord2Vec/main04.ipynb diff --git a/05_Classification/main.ipynb b/05_Classification/main05.ipynb similarity index 99% rename from 05_Classification/main.ipynb rename to 05_Classification/main05.ipynb index 523b00f..e39a103 100644 --- a/05_Classification/main.ipynb +++ b/05_Classification/main05.ipynb @@ -2904,7 +2904,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.6.5" + "version": "3.6.4" } }, "nbformat": 4, diff --git a/06_XgbCV/main.ipynb b/06_XgbCV/main.ipynb deleted file mode 100644 index b3c6635..0000000 --- a/06_XgbCV/main.ipynb +++ /dev/null @@ -1,3194 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# XGB + CV" - ] - }, - { - "cell_type": "code", - "execution_count": 18, - "metadata": {}, - "outputs": [], - "source": [ - "import matplotlib.pyplot as plt\n", - "from planar_utils import plot_decision_boundary, sigmoid, load_planar_dataset, load_extra_datasets\n", - "import numpy as np\n", - "import pandas as pd\n", - "import os\n", - "from collections import Counter \n", - "\n", - "from sklearn.neighbors import KNeighborsClassifier ## KNN\n", - "from sklearn.linear_model import LogisticRegressionCV ## logistic regression\n", - "from sklearn.tree import DecisionTreeClassifier ## decision tree\n", - "from sklearn.svm import SVC ## SVM\n", - "\n", - "from sklearn.tree import DecisionTreeClassifier ## decision tree\n", - "from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier\n", - "from xgboost import XGBClassifier\n", - "\n", - "import math\n", - "import string\n", - "import re\n", - "\n", - "import xgboost\n", - "\n", - "from preprocess import preprocess" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "collapsed": true - }, - "source": [ - "# 鐵達尼號資料集" - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "metadata": { - "scrolled": true - }, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
PassengerIdSurvivedPclassSexAgeSibSpParchTicketCabinEmbarkedHas_CabinAge_CatFare_log2Fare_CatName_LengthName_With_Special_CharFamily_SizeTitle
0103122.010200012.857981023011
1211038.010532126.155492551113
2313026.000700012.986411022002
3411035.010130125.730640444113
4503135.000100023.008989024001
\n", - "
" - ], - "text/plain": [ - " PassengerId Survived Pclass Sex Age SibSp Parch Ticket Cabin \\\n", - "0 1 0 3 1 22.0 1 0 2 0 \n", - "1 2 1 1 0 38.0 1 0 5 3 \n", - "2 3 1 3 0 26.0 0 0 7 0 \n", - "3 4 1 1 0 35.0 1 0 1 3 \n", - "4 5 0 3 1 35.0 0 0 1 0 \n", - "\n", - " Embarked Has_Cabin Age_Cat Fare_log2 Fare_Cat Name_Length \\\n", - "0 0 0 1 2.857981 0 23 \n", - "1 2 1 2 6.155492 5 51 \n", - "2 0 0 1 2.986411 0 22 \n", - "3 0 1 2 5.730640 4 44 \n", - "4 0 0 2 3.008989 0 24 \n", - "\n", - " Name_With_Special_Char Family_Size Title \n", - "0 0 1 1 \n", - "1 1 1 3 \n", - "2 0 0 2 \n", - "3 1 1 3 \n", - "4 0 0 1 " - ] - }, - "execution_count": 2, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df = pd.read_csv('train.csv')\n", - "df = preprocess(df)\n", - "df.head()" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": {}, - "outputs": [], - "source": [ - "X = df[['PassengerId', 'Pclass', 'Sex', 'Age', 'SibSp', 'Parch',\n", - " 'Ticket', 'Cabin', 'Embarked', 'Has_Cabin', 'Age_Cat', 'Fare_log2',\n", - " 'Fare_Cat', 'Name_Length', 'Name_With_Special_Char', 'Family_Size',\n", - " 'Title']].values\n", - "Y = df['Survived'].values" - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "(623, 17)\n", - "(268, 17)\n", - "(623,)\n", - "(268,)\n" - ] - } - ], - "source": [ - "from sklearn.model_selection import train_test_split\n", - "\n", - "X_train, X_valid, Y_train, Y_valid = train_test_split(X, Y, test_size =0.3, random_state=1212) ## 一般如果測試資料集超過1000筆就可以了，所以比率不會設這麼高\n", - "print(X_train.shape) ## (445, 17)\n", - "print(X_valid.shape) ## (446, 17) \n", - "print(Y_train.shape) ## (445,)\n", - "print(Y_valid.shape) ## (446,)" - ] - }, - { - "cell_type": "code", - "execution_count": 6, - "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "c:\\users\\user\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\sklearn\\svm\\base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n", - " \"avoid this warning.\", FutureWarning)\n", - "c:\\users\\user\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\sklearn\\ensemble\\forest.py:248: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n", - " \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "SVM: 0.6455223880597015\n", - "DecisionTree: 0.7649253731343284\n", - "RandomForest: 0.8283582089552238\n", - "AdaBoost: 0.7910447761194029\n", - "XGB: 0.8432835820895522\n" - ] - } - ], - "source": [ - "def get_accuracy(clf):\n", - " clf = clf()\n", - " clf = clf.fit(X_train, Y_train)\n", - " y_pred = clf.predict(X_valid)\n", - " return (str(sum(Y_valid == y_pred)/Y_valid.shape[0]))\n", - "\n", - "print('SVM: ', get_accuracy(SVC))\n", - "print('DecisionTree: ', get_accuracy(DecisionTreeClassifier))\n", - "print('RandomForest: ', get_accuracy(RandomForestClassifier))\n", - "print('AdaBoost: ', get_accuracy(AdaBoostClassifier)) ## Boosting的演算法\n", - "print('XGB: ', get_accuracy(XGBClassifier))\n", - "\n", - "# SVM: 0.609865470852\n", - "# DecisionTree: 0.764573991031\n", - "# RandomForest: 0.795964125561\n", - "# AdaBoost: 0.784753363229\n", - "# XGB: 0.80269058296" - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "metadata": { - "scrolled": true - }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "[08:03:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", - "Accuracy: 0.8059701492537313\n" - ] - } - ], - "source": [ - "# Set our parameters for xgboost\n", - "params = {}\n", - "params['objective'] = 'binary:logistic'\n", - "params['eval_metric'] = 'logloss'\n", - "params['eta'] = 0.04\n", - "params['max_depth'] = 3\n", - "params['learning_rate'] = 0.001\n", - "\n", - "d_train = xgboost.DMatrix(X_train, label=Y_train)\n", - "d_valid = xgboost.DMatrix(X_valid, label=Y_valid)\n", - "\n", - "watchlist = [(d_train, 'train'), (d_valid, 'valid')]\n", - "\n", - "bst = xgboost.train(params, d_train, 100, watchlist, early_stopping_rounds=100, verbose_eval=0)\n", - "y_pred = bst.predict(xgboost.DMatrix(X_valid))\n", - "print(\"Accuracy: \", str(sum(Y_valid == (y_pred > 0.5))/Y_valid.shape[0]))\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "collapsed": true - }, - "source": [ - "# 房價資料集" - ] - }, - { - "cell_type": "code", - "execution_count": 12, - "metadata": {}, - "outputs": [], - "source": [ - "import urllib.request\n", - "if 'df_realestate_processed.csv' not in os.listdir():\n", - " url = 'https://s3.amazonaws.com/datasets-jeremy/df_realestate_processed.csv' \n", - " urllib.request.urlretrieve(url, 'df_realestate_processed.csv')\n", - " \n", - "# processed\n", - "path = \"df_realestate_processed.csv\"\n", - "df_realestate_processed = pd.read_csv(path)\n", - "X = df_realestate_processed.drop([\"price_per_meter\", \"total_price\"], axis=1)\n", - "Y = df_realestate_processed['total_price']\n" - ] - }, - { - "cell_type": "code", - "execution_count": 13, - "metadata": {}, - "outputs": [], - "source": [ - "X_train = X.iloc[:-1000]\n", - "Y_train = Y.iloc[:-1000]\n", - "Y_train = np.log(Y_train) \n", - "\n", - "X_valid = X.iloc[-1000:]\n", - "Y_valid = Y.iloc[-1000:]\n", - "Y_valid = np.log(Y_valid) " - ] - }, - { - "cell_type": "code", - "execution_count": 14, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:25:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[0]\ttrain-rmse:16.1369\tvalid-rmse:15.9434\n", - "Multiple eval metrics have been passed: 'valid-rmse' will be used for early stopping.\n", - "\n", - "Will train until valid-rmse hasn't improved in 100 rounds.\n", - "[08:25:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 22 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[10]\ttrain-rmse:14.5951\tvalid-rmse:14.4244\n", - "[08:25:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:25:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[20]\ttrain-rmse:13.2008\tvalid-rmse:13.0514\n", - "[08:26:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 22 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n", - "[30]\ttrain-rmse:11.9398\tvalid-rmse:11.8096\n", - "[08:26:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 0 pruned nodes, max_depth=5\n", - "[40]\ttrain-rmse:10.7995\tvalid-rmse:10.6865\n", - "[08:26:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=4\n", - "[08:26:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=5\n", - "[50]\ttrain-rmse:9.76818\tvalid-rmse:9.67152\n", - "[08:26:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:26:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=4\n", - "[60]\ttrain-rmse:8.83554\tvalid-rmse:8.75437\n", - "[08:26:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 36 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 36 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 0 pruned nodes, max_depth=5\n", - "[70]\ttrain-rmse:7.99212\tvalid-rmse:7.92454\n", - "[08:26:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 36 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 40 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 36 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 0 pruned nodes, max_depth=5\n", - "[80]\ttrain-rmse:7.22941\tvalid-rmse:7.17414\n", - "[08:26:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 36 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 42 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 42 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 42 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 36 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:26:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:26:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=6\n", - "[90]\ttrain-rmse:6.53968\tvalid-rmse:6.49537\n", - "[08:27:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 44 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 36 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:27:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 44 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 36 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 38 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 42 extra nodes, 0 pruned nodes, max_depth=6\n", - "[100]\ttrain-rmse:5.91598\tvalid-rmse:5.88217\n", - "[08:27:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 44 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 36 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:27:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 42 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 46 extra nodes, 0 pruned nodes, max_depth=6\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:27:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 40 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 40 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 42 extra nodes, 0 pruned nodes, max_depth=5\n", - "[08:27:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 44 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 46 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 42 extra nodes, 0 pruned nodes, max_depth=6\n", - "[110]\ttrain-rmse:5.35201\tvalid-rmse:5.32788\n", - "[08:27:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 44 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 46 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 52 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 46 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 46 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 44 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 44 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 42 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 52 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 54 extra nodes, 0 pruned nodes, max_depth=6\n", - "[120]\ttrain-rmse:4.84205\tvalid-rmse:4.82646\n", - "[08:27:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 48 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 58 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 46 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 46 extra nodes, 0 pruned nodes, max_depth=6\n", - "[08:27:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 56 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 56 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 54 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 56 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 54 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 56 extra nodes, 0 pruned nodes, max_depth=7\n", - "[130]\ttrain-rmse:4.38097\tvalid-rmse:4.3728\n", - "[08:27:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 56 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 60 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 54 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 62 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 60 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 66 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 52 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 74 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 64 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 66 extra nodes, 0 pruned nodes, max_depth=7\n", - "[140]\ttrain-rmse:3.96408\tvalid-rmse:3.96239\n", - "[08:27:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 70 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 64 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 64 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 68 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 64 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 70 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 66 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 72 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 74 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 64 extra nodes, 0 pruned nodes, max_depth=7\n", - "[150]\ttrain-rmse:3.58718\tvalid-rmse:3.59095\n", - "[08:27:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 80 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 68 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 72 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 88 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:27:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 80 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:27:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 86 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 82 extra nodes, 0 pruned nodes, max_depth=7\n", - "[160]\ttrain-rmse:3.24645\tvalid-rmse:3.25507\n", - "[08:28:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 92 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 90 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 88 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 86 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 94 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 96 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 82 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[170]\ttrain-rmse:2.93845\tvalid-rmse:2.95125\n", - "[08:28:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 82 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 98 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 96 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 94 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 88 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[180]\ttrain-rmse:2.66005\tvalid-rmse:2.67638\n", - "[08:28:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 96 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[190]\ttrain-rmse:2.40845\tvalid-rmse:2.42759\n", - "[08:28:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[200]\ttrain-rmse:2.18109\tvalid-rmse:2.20289\n", - "[08:28:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[210]\ttrain-rmse:1.97567\tvalid-rmse:2.0001\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:28:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[220]\ttrain-rmse:1.79013\tvalid-rmse:1.81692\n", - "[08:28:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[230]\ttrain-rmse:1.62258\tvalid-rmse:1.65169\n", - "[08:28:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:28:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[240]\ttrain-rmse:1.47132\tvalid-rmse:1.50286\n", - "[08:29:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[250]\ttrain-rmse:1.3348\tvalid-rmse:1.36886\n", - "[08:29:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[260]\ttrain-rmse:1.21168\tvalid-rmse:1.24834\n", - "[08:29:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:29:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[270]\ttrain-rmse:1.10068\tvalid-rmse:1.13999\n", - "[08:29:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[280]\ttrain-rmse:1.00069\tvalid-rmse:1.04218\n", - "[08:29:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[290]\ttrain-rmse:0.910688\tvalid-rmse:0.954809\n", - "[08:29:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:29:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[300]\ttrain-rmse:0.829754\tvalid-rmse:0.876668\n", - "[08:29:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[310]\ttrain-rmse:0.757038\tvalid-rmse:0.806862\n", - "[08:30:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:30:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[320]\ttrain-rmse:0.691817\tvalid-rmse:0.744808\n", - "[08:30:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 214 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[330]\ttrain-rmse:0.633404\tvalid-rmse:0.689832\n", - "[08:30:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 214 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 214 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[340]\ttrain-rmse:0.581149\tvalid-rmse:0.640921\n", - "[08:30:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[350]\ttrain-rmse:0.534521\tvalid-rmse:0.597822\n", - "[08:30:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 214 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[360]\ttrain-rmse:0.493021\tvalid-rmse:0.559893\n", - "[08:30:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 214 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:30:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[370]\ttrain-rmse:0.45619\tvalid-rmse:0.526247\n", - "[08:30:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:30:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[380]\ttrain-rmse:0.423568\tvalid-rmse:0.497081\n", - "[08:31:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 228 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 228 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[390]\ttrain-rmse:0.394781\tvalid-rmse:0.471699\n", - "[08:31:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 232 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 214 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 228 extra nodes, 0 pruned nodes, max_depth=7\n", - "[400]\ttrain-rmse:0.369466\tvalid-rmse:0.449716\n", - "[08:31:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 228 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 226 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 226 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 226 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[410]\ttrain-rmse:0.347277\tvalid-rmse:0.430813\n", - "[08:31:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 218 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 218 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 228 extra nodes, 0 pruned nodes, max_depth=7\n", - "[420]\ttrain-rmse:0.327938\tvalid-rmse:0.414746\n", - "[08:31:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 218 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:31:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 214 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 224 extra nodes, 0 pruned nodes, max_depth=7\n", - "[430]\ttrain-rmse:0.311128\tvalid-rmse:0.400968\n", - "[08:31:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 224 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[440]\ttrain-rmse:0.296527\tvalid-rmse:0.389149\n", - "[08:31:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 218 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 222 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 222 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 222 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:31:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 222 extra nodes, 0 pruned nodes, max_depth=7\n", - "[450]\ttrain-rmse:0.283912\tvalid-rmse:0.379036\n", - "[08:32:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 228 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[460]\ttrain-rmse:0.273086\tvalid-rmse:0.370692\n", - "[08:32:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 218 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 228 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 216 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 218 extra nodes, 0 pruned nodes, max_depth=7\n", - "[470]\ttrain-rmse:0.263794\tvalid-rmse:0.363678\n", - "[08:32:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 216 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:32:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[480]\ttrain-rmse:0.255845\tvalid-rmse:0.357647\n", - "[08:32:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[490]\ttrain-rmse:0.249099\tvalid-rmse:0.352454\n", - "[08:32:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[500]\ttrain-rmse:0.243296\tvalid-rmse:0.348231\n", - "[08:32:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[510]\ttrain-rmse:0.238395\tvalid-rmse:0.344567\n", - "[08:32:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:32:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 222 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[520]\ttrain-rmse:0.234227\tvalid-rmse:0.341425\n", - "[08:33:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 216 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:33:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[530]\ttrain-rmse:0.230658\tvalid-rmse:0.338733\n", - "[08:33:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 214 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[540]\ttrain-rmse:0.227576\tvalid-rmse:0.336438\n", - "[08:33:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 214 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 222 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[550]\ttrain-rmse:0.224917\tvalid-rmse:0.33455\n", - "[08:33:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 224 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[560]\ttrain-rmse:0.222633\tvalid-rmse:0.332905\n", - "[08:33:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 218 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[570]\ttrain-rmse:0.220691\tvalid-rmse:0.33156\n", - "[08:33:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:33:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[580]\ttrain-rmse:0.218983\tvalid-rmse:0.330504\n", - "[08:34:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:34:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[590]\ttrain-rmse:0.217472\tvalid-rmse:0.329516\n", - "[08:34:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[600]\ttrain-rmse:0.216157\tvalid-rmse:0.328659\n", - "[08:34:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[610]\ttrain-rmse:0.215\tvalid-rmse:0.327922\n", - "[08:34:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 216 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[620]\ttrain-rmse:0.213951\tvalid-rmse:0.327293\n", - "[08:34:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[630]\ttrain-rmse:0.213033\tvalid-rmse:0.326807\n", - "[08:34:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 216 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:34:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[640]\ttrain-rmse:0.212183\tvalid-rmse:0.326362\n", - "[08:34:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:34:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[650]\ttrain-rmse:0.211424\tvalid-rmse:0.325891\n", - "[08:35:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[660]\ttrain-rmse:0.210771\tvalid-rmse:0.325551\n", - "[08:35:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[670]\ttrain-rmse:0.210114\tvalid-rmse:0.325268\n", - "[08:35:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[680]\ttrain-rmse:0.209548\tvalid-rmse:0.325044\n", - "[08:35:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:35:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[690]\ttrain-rmse:0.208988\tvalid-rmse:0.324729\n", - "[08:35:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[700]\ttrain-rmse:0.208486\tvalid-rmse:0.324464\n", - "[08:35:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[710]\ttrain-rmse:0.20802\tvalid-rmse:0.324186\n", - "[08:35:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:35:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[720]\ttrain-rmse:0.207545\tvalid-rmse:0.323965\n", - "[08:36:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[730]\ttrain-rmse:0.207132\tvalid-rmse:0.323707\n", - "[08:36:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[740]\ttrain-rmse:0.206738\tvalid-rmse:0.32355\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:36:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[750]\ttrain-rmse:0.206327\tvalid-rmse:0.323482\n", - "[08:36:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[760]\ttrain-rmse:0.20594\tvalid-rmse:0.323252\n", - "[08:36:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[770]\ttrain-rmse:0.205567\tvalid-rmse:0.323104\n", - "[08:36:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[780]\ttrain-rmse:0.205221\tvalid-rmse:0.323046\n", - "[08:36:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[790]\ttrain-rmse:0.204867\tvalid-rmse:0.322917\n", - "[08:36:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:36:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:37:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[800]\ttrain-rmse:0.204559\tvalid-rmse:0.322805\n", - "[08:37:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[810]\ttrain-rmse:0.204237\tvalid-rmse:0.322725\n", - "[08:37:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[820]\ttrain-rmse:0.203925\tvalid-rmse:0.322572\n", - "[08:37:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 94 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[830]\ttrain-rmse:0.203625\tvalid-rmse:0.322487\n", - "[08:37:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[840]\ttrain-rmse:0.20333\tvalid-rmse:0.322384\n", - "[08:37:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:37:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[850]\ttrain-rmse:0.203038\tvalid-rmse:0.322271\n", - "[08:37:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[860]\ttrain-rmse:0.202768\tvalid-rmse:0.322175\n", - "[08:37:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:37:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[870]\ttrain-rmse:0.202511\tvalid-rmse:0.322117\n", - "[08:38:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[880]\ttrain-rmse:0.202248\tvalid-rmse:0.322034\n", - "[08:38:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[890]\ttrain-rmse:0.201956\tvalid-rmse:0.321935\n", - "[08:38:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:38:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[900]\ttrain-rmse:0.201713\tvalid-rmse:0.321881\n", - "[08:38:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[910]\ttrain-rmse:0.201467\tvalid-rmse:0.321846\n", - "[08:38:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[920]\ttrain-rmse:0.201207\tvalid-rmse:0.321795\n", - "[08:38:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[930]\ttrain-rmse:0.200889\tvalid-rmse:0.321767\n", - "[08:38:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[940]\ttrain-rmse:0.200634\tvalid-rmse:0.321664\n", - "[08:38:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:38:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[950]\ttrain-rmse:0.200377\tvalid-rmse:0.321587\n", - "[08:39:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:39:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[960]\ttrain-rmse:0.20009\tvalid-rmse:0.321546\n", - "[08:39:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[970]\ttrain-rmse:0.19981\tvalid-rmse:0.321571\n", - "[08:39:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[980]\ttrain-rmse:0.199496\tvalid-rmse:0.321525\n", - "[08:39:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[990]\ttrain-rmse:0.199294\tvalid-rmse:0.32142\n", - "[08:39:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1000]\ttrain-rmse:0.199037\tvalid-rmse:0.32131\n", - "[08:39:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:39:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1010]\ttrain-rmse:0.198804\tvalid-rmse:0.321233\n", - "[08:39:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:39:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1020]\ttrain-rmse:0.198536\tvalid-rmse:0.321163\n", - "[08:40:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1030]\ttrain-rmse:0.198282\tvalid-rmse:0.321071\n", - "[08:40:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1040]\ttrain-rmse:0.198088\tvalid-rmse:0.321001\n", - "[08:40:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1050]\ttrain-rmse:0.197848\tvalid-rmse:0.32099\n", - "[08:40:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 218 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:40:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1060]\ttrain-rmse:0.197558\tvalid-rmse:0.320927\n", - "[08:40:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1070]\ttrain-rmse:0.197362\tvalid-rmse:0.320917\n", - "[08:40:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1080]\ttrain-rmse:0.197084\tvalid-rmse:0.320766\n", - "[08:40:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1090]\ttrain-rmse:0.196893\tvalid-rmse:0.320726\n", - "[08:40:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:40:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1100]\ttrain-rmse:0.196646\tvalid-rmse:0.320685\n", - "[08:41:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 216 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1110]\ttrain-rmse:0.196377\tvalid-rmse:0.320658\n", - "[08:41:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:41:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1120]\ttrain-rmse:0.196177\tvalid-rmse:0.32062\n", - "[08:41:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1130]\ttrain-rmse:0.195956\tvalid-rmse:0.320642\n", - "[08:41:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1140]\ttrain-rmse:0.195763\tvalid-rmse:0.320604\n", - "[08:41:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1150]\ttrain-rmse:0.195568\tvalid-rmse:0.320561\n", - "[08:41:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1160]\ttrain-rmse:0.195371\tvalid-rmse:0.320545\n", - "[08:41:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:41:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:41:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1170]\ttrain-rmse:0.195172\tvalid-rmse:0.320559\n", - "[08:42:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1180]\ttrain-rmse:0.194987\tvalid-rmse:0.320549\n", - "[08:42:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1190]\ttrain-rmse:0.194826\tvalid-rmse:0.320573\n", - "[08:42:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 224 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1200]\ttrain-rmse:0.194568\tvalid-rmse:0.320569\n", - "[08:42:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 224 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1210]\ttrain-rmse:0.194336\tvalid-rmse:0.320548\n", - "[08:42:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:42:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1220]\ttrain-rmse:0.194097\tvalid-rmse:0.320555\n", - "[08:42:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1230]\ttrain-rmse:0.193876\tvalid-rmse:0.320523\n", - "[08:42:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 226 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 220 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1240]\ttrain-rmse:0.193635\tvalid-rmse:0.320506\n", - "[08:42:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:42:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 96 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 98 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1250]\ttrain-rmse:0.193438\tvalid-rmse:0.320466\n", - "[08:43:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1260]\ttrain-rmse:0.193257\tvalid-rmse:0.320442\n", - "[08:43:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1270]\ttrain-rmse:0.193034\tvalid-rmse:0.320454\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:43:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1280]\ttrain-rmse:0.192855\tvalid-rmse:0.320536\n", - "[08:43:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1290]\ttrain-rmse:0.19264\tvalid-rmse:0.32052\n", - "[08:43:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1300]\ttrain-rmse:0.192393\tvalid-rmse:0.320443\n", - "[08:43:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 92 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1310]\ttrain-rmse:0.19222\tvalid-rmse:0.320421\n", - "[08:43:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:43:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1320]\ttrain-rmse:0.192018\tvalid-rmse:0.32041\n", - "[08:44:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 216 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:44:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1330]\ttrain-rmse:0.191789\tvalid-rmse:0.320355\n", - "[08:44:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1340]\ttrain-rmse:0.191585\tvalid-rmse:0.320382\n", - "[08:44:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1350]\ttrain-rmse:0.191343\tvalid-rmse:0.320379\n", - "[08:44:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1360]\ttrain-rmse:0.191149\tvalid-rmse:0.320392\n", - "[08:44:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1370]\ttrain-rmse:0.190934\tvalid-rmse:0.320408\n", - "[08:44:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:44:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 90 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1380]\ttrain-rmse:0.190746\tvalid-rmse:0.320475\n", - "[08:44:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:44:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 86 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1390]\ttrain-rmse:0.190567\tvalid-rmse:0.320481\n", - "[08:45:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 88 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1400]\ttrain-rmse:0.190375\tvalid-rmse:0.320499\n", - "[08:45:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1410]\ttrain-rmse:0.190208\tvalid-rmse:0.320436\n", - "[08:45:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1420]\ttrain-rmse:0.190019\tvalid-rmse:0.320341\n", - "[08:45:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:45:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1430]\ttrain-rmse:0.189834\tvalid-rmse:0.320335\n", - "[08:45:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 210 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 82 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1440]\ttrain-rmse:0.189641\tvalid-rmse:0.320305\n", - "[08:45:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 90 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 78 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1450]\ttrain-rmse:0.189484\tvalid-rmse:0.320324\n", - "[08:45:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:45:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 92 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1460]\ttrain-rmse:0.189305\tvalid-rmse:0.320339\n", - "[08:46:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 80 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1470]\ttrain-rmse:0.189157\tvalid-rmse:0.320319\n", - "[08:46:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1480]\ttrain-rmse:0.188966\tvalid-rmse:0.320297\n", - "[08:46:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:46:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1490]\ttrain-rmse:0.188728\tvalid-rmse:0.320218\n", - "[08:46:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 82 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1500]\ttrain-rmse:0.188559\tvalid-rmse:0.320205\n", - "[08:46:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1510]\ttrain-rmse:0.188356\tvalid-rmse:0.320091\n", - "[08:46:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1520]\ttrain-rmse:0.188187\tvalid-rmse:0.31998\n", - "[08:46:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:46:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 212 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 82 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1530]\ttrain-rmse:0.188007\tvalid-rmse:0.319952\n", - "[08:47:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:47:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1540]\ttrain-rmse:0.187759\tvalid-rmse:0.319893\n", - "[08:47:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 198 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 200 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1550]\ttrain-rmse:0.187562\tvalid-rmse:0.319846\n", - "[08:47:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1560]\ttrain-rmse:0.187419\tvalid-rmse:0.319863\n", - "[08:47:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 228 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 98 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1570]\ttrain-rmse:0.187216\tvalid-rmse:0.319856\n", - "[08:47:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1580]\ttrain-rmse:0.187026\tvalid-rmse:0.319856\n", - "[08:47:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 86 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:47:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1590]\ttrain-rmse:0.186847\tvalid-rmse:0.31982\n", - "[08:47:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 90 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:47:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1600]\ttrain-rmse:0.186696\tvalid-rmse:0.319861\n", - "[08:48:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 88 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 90 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1610]\ttrain-rmse:0.186562\tvalid-rmse:0.319866\n", - "[08:48:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 96 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1620]\ttrain-rmse:0.18636\tvalid-rmse:0.31987\n", - "[08:48:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1630]\ttrain-rmse:0.186196\tvalid-rmse:0.319858\n", - "[08:48:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 86 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1640]\ttrain-rmse:0.186031\tvalid-rmse:0.319854\n", - "[08:48:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:48:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1650]\ttrain-rmse:0.185857\tvalid-rmse:0.31983\n", - "[08:48:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 78 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1660]\ttrain-rmse:0.185698\tvalid-rmse:0.319803\n", - "[08:48:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1670]\ttrain-rmse:0.18554\tvalid-rmse:0.319814\n", - "[08:48:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:48:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1680]\ttrain-rmse:0.185382\tvalid-rmse:0.319803\n", - "[08:49:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 202 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 94 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1690]\ttrain-rmse:0.185224\tvalid-rmse:0.319781\n", - "[08:49:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:49:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1700]\ttrain-rmse:0.185029\tvalid-rmse:0.319754\n", - "[08:49:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 72 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1710]\ttrain-rmse:0.184869\tvalid-rmse:0.319758\n", - "[08:49:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1720]\ttrain-rmse:0.184677\tvalid-rmse:0.319739\n", - "[08:49:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 106 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 96 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1730]\ttrain-rmse:0.184528\tvalid-rmse:0.319722\n", - "[08:49:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1740]\ttrain-rmse:0.184346\tvalid-rmse:0.319715\n", - "[08:49:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 96 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:49:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 86 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 208 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:50:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 204 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 94 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1750]\ttrain-rmse:0.184197\tvalid-rmse:0.319715\n", - "[08:50:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 226 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1760]\ttrain-rmse:0.184034\tvalid-rmse:0.319684\n", - "[08:50:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1770]\ttrain-rmse:0.183875\tvalid-rmse:0.319624\n", - "[08:50:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 82 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 194 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1780]\ttrain-rmse:0.183707\tvalid-rmse:0.319614\n", - "[08:50:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 92 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 92 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1790]\ttrain-rmse:0.183557\tvalid-rmse:0.319616\n", - "[08:50:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 72 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 98 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 82 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1800]\ttrain-rmse:0.18342\tvalid-rmse:0.319588\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:50:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 88 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 98 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 126 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1810]\ttrain-rmse:0.183292\tvalid-rmse:0.319633\n", - "[08:50:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 82 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:50:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1820]\ttrain-rmse:0.183151\tvalid-rmse:0.319622\n", - "[08:51:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 128 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 72 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1830]\ttrain-rmse:0.183006\tvalid-rmse:0.319644\n", - "[08:51:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 178 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 110 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 196 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1840]\ttrain-rmse:0.182841\tvalid-rmse:0.31969\n", - "[08:51:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 98 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 78 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1850]\ttrain-rmse:0.1827\tvalid-rmse:0.319712\n", - "[08:51:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:51:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 100 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 96 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1860]\ttrain-rmse:0.182529\tvalid-rmse:0.319707\n", - "[08:51:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1870]\ttrain-rmse:0.182386\tvalid-rmse:0.319712\n", - "[08:51:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 182 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1880]\ttrain-rmse:0.182224\tvalid-rmse:0.319641\n", - "[08:51:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 98 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 162 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1890]\ttrain-rmse:0.182075\tvalid-rmse:0.319585\n", - "[08:51:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 72 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:51:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 98 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 74 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 170 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1900]\ttrain-rmse:0.181944\tvalid-rmse:0.319568\n", - "[08:52:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 132 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:52:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 108 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 184 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1910]\ttrain-rmse:0.181778\tvalid-rmse:0.319574\n", - "[08:52:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 192 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 78 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1920]\ttrain-rmse:0.181613\tvalid-rmse:0.319581\n", - "[08:52:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 92 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 172 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 168 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 146 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 84 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1930]\ttrain-rmse:0.181479\tvalid-rmse:0.319602\n", - "[08:52:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 190 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 112 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 96 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 138 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 206 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 90 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 88 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1940]\ttrain-rmse:0.181341\tvalid-rmse:0.319618\n", - "[08:52:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 116 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 150 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 142 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 104 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1950]\ttrain-rmse:0.181202\tvalid-rmse:0.319638\n", - "[08:52:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 90 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 148 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 158 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 134 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 80 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 186 extra nodes, 0 pruned nodes, max_depth=7\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[08:52:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 176 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1960]\ttrain-rmse:0.181067\tvalid-rmse:0.319674\n", - "[08:52:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 122 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 88 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 58 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 72 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 114 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:52:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1970]\ttrain-rmse:0.180982\tvalid-rmse:0.31972\n", - "[08:53:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 124 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 188 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 156 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 88 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 58 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 82 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 130 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 136 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1980]\ttrain-rmse:0.180867\tvalid-rmse:0.319684\n", - "[08:53:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 86 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 166 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 120 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 160 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 154 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "[1990]\ttrain-rmse:0.180719\tvalid-rmse:0.319726\n", - "[08:53:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 118 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 180 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 102 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 98 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 174 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 144 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 164 extra nodes, 0 pruned nodes, max_depth=7\n", - "[08:53:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 140 extra nodes, 0 pruned nodes, max_depth=7\n", - "[2000]\ttrain-rmse:0.180576\tvalid-rmse:0.319734\n", - "[08:53:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 152 extra nodes, 0 pruned nodes, max_depth=7\n", - "Stopping. Best iteration:\n", - "[1901]\ttrain-rmse:0.181926\tvalid-rmse:0.319563\n", - "\n" - ] - } - ], - "source": [ - "# Set our parameters for xgboost\n", - "params = {}\n", - "params['objective'] = 'reg:linear'\n", - "params['eval_metric'] = 'rmse'\n", - "params['eta'] = 0.01\n", - "params['max_depth'] = 5\n", - "\n", - "d_train = xgboost.DMatrix(X_train, label=Y_train)\n", - "d_valid = xgboost.DMatrix(X_valid, label=Y_valid)\n", - "\n", - "watchlist = [(d_train, 'train'), (d_valid, 'valid')]\n", - "\n", - "bst = xgboost.train(params, d_train, 3000, watchlist, early_stopping_rounds=50, verbose_eval=10)\n", - "Y_pred = bst.predict(xgboost.DMatrix(X_valid))\n" - ] - }, - { - "cell_type": "code", - "execution_count": 15, - "metadata": { - "scrolled": true - }, - "outputs": [ - { - "data": { - "image/png": "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\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], - "source": [ - "xgboost.plot_importance(bst, max_num_features=10)\n", - "plt.show()" - ] - }, - { - "cell_type": "code", - "execution_count": 16, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
predicttrutherror
8101331979.375400002.562.329927
5371273172.125470004.001.708854
8091322311.000500003.201.644605
8121481586.250689993.791.147246
8111481586.250739999.241.002146
\n", - "
" - ], - "text/plain": [ - " predict truth error\n", - "810 1331979.375 400002.56 2.329927\n", - "537 1273172.125 470004.00 1.708854\n", - "809 1322311.000 500003.20 1.644605\n", - "812 1481586.250 689993.79 1.147246\n", - "811 1481586.250 739999.24 1.002146" - ] - }, - "execution_count": 16, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_result = pd.DataFrame()\n", - "Y_pred = bst.predict(xgboost.DMatrix(X_valid))\n", - "df_result['predict'] = np.exp(Y_pred)\n", - "df_result['truth'] = np.exp((list(Y_valid)))\n", - "df_result.sort_values('truth', inplace=True)\n", - "df_result['error'] = df_result.apply(lambda x:np.abs(x['predict'] - x['truth']) / x['truth'], axis=1)\n", - "df_result.head()" - ] - }, - { - "cell_type": "code", - "execution_count": 19, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "Counter({Interval(0.267, 0.356, closed='right'): 78,\n", - " Interval(-0.000889, 0.089, closed='right'): 356,\n", - " Interval(0.712, 0.801, closed='right'): 5,\n", - " Interval(0.089, 0.178, closed='right'): 252,\n", - " Interval(0.445, 0.534, closed='right'): 44,\n", - " Interval(0.356, 0.445, closed='right'): 53,\n", - " Interval(0.801, 0.89, closed='right'): 9,\n", - " Interval(0.534, 0.623, closed='right'): 20,\n", - " Interval(0.178, 0.267, closed='right'): 153,\n", - " Interval(0.623, 0.712, closed='right'): 15})" - ] - }, - "execution_count": 19, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "Counter(pd.cut(df_result.loc[df_result['error'] < 1, 'error'], bins=10))\n" - ] - }, - { - "cell_type": "code", - "execution_count": 20, - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], - "source": [ - "df_result.loc[df_result['error'] < 1, 'error'].plot('hist')\n", - "\n", - "plt.show()" - ] - }, - { - "cell_type": "code", - "execution_count": 21, - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], - "source": [ - "plt.scatter(range(len(df_result)), df_result['predict'].values, color='black', s=0.5)\n", - "\n", - "plt.scatter(range(len(df_result)), df_result['truth'].values, color='red', s=0.5)\n", - "\n", - "plt.show()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.6.4" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/06_XgbCV/main06.ipynb b/06_XgbCV/main06.ipynb new file mode 100644 index 0000000..0b919f3 --- /dev/null +++ b/06_XgbCV/main06.ipynb @@ -0,0 +1,1411 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# XGB + CV" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [], + "source": [ + "import matplotlib.pyplot as plt\n", + "from planar_utils import plot_decision_boundary, sigmoid, load_planar_dataset, load_extra_datasets\n", + "import numpy as np\n", + "import pandas as pd\n", + "import os\n", + "from collections import Counter \n", + "\n", + "from sklearn.neighbors import KNeighborsClassifier ## KNN\n", + "from sklearn.linear_model import LogisticRegressionCV ## logistic regression\n", + "from sklearn.tree import DecisionTreeClassifier ## decision tree\n", + "from sklearn.svm import SVC ## SVM\n", + "\n", + "from sklearn.tree import DecisionTreeClassifier ## decision tree\n", + "from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier\n", + "from xgboost import XGBClassifier\n", + "\n", + "import math\n", + "import string\n", + "import re\n", + "\n", + "import xgboost\n", + "\n", + "from preprocess import preprocess" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": true + }, + "source": [ + "# 鐵達尼號資料集" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
PassengerIdSurvivedPclassSexAgeSibSpParchTicketCabinEmbarkedHas_CabinAge_CatFare_log2Fare_CatName_LengthName_With_Special_CharFamily_SizeTitle
0103122.010200012.857981023011
1211038.010532126.155492551113
2313026.000700012.986411022002
3411035.010130125.730640444113
4503135.000100023.008989024001
\n", + "
" + ], + "text/plain": [ + " PassengerId Survived Pclass Sex Age SibSp Parch Ticket Cabin \\\n", + "0 1 0 3 1 22.0 1 0 2 0 \n", + "1 2 1 1 0 38.0 1 0 5 3 \n", + "2 3 1 3 0 26.0 0 0 7 0 \n", + "3 4 1 1 0 35.0 1 0 1 3 \n", + "4 5 0 3 1 35.0 0 0 1 0 \n", + "\n", + " Embarked Has_Cabin Age_Cat Fare_log2 Fare_Cat Name_Length \\\n", + "0 0 0 1 2.857981 0 23 \n", + "1 2 1 2 6.155492 5 51 \n", + "2 0 0 1 2.986411 0 22 \n", + "3 0 1 2 5.730640 4 44 \n", + "4 0 0 2 3.008989 0 24 \n", + "\n", + " Name_With_Special_Char Family_Size Title \n", + "0 0 1 1 \n", + "1 1 1 3 \n", + "2 0 0 2 \n", + "3 1 1 3 \n", + "4 0 0 1 " + ] + }, + "execution_count": 3, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df = pd.read_csv('train.csv')\n", + "df = preprocess(df)\n", + "df.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [], + "source": [ + "X = df[['PassengerId', 'Pclass', 'Sex', 'Age', 'SibSp', 'Parch',\n", + " 'Ticket', 'Cabin', 'Embarked', 'Has_Cabin', 'Age_Cat', 'Fare_log2',\n", + " 'Fare_Cat', 'Name_Length', 'Name_With_Special_Char', 'Family_Size',\n", + " 'Title']].values\n", + "Y = df['Survived'].values" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "(623, 17)\n", + "(268, 17)\n", + "(623,)\n", + "(268,)\n" + ] + } + ], + "source": [ + "from sklearn.model_selection import train_test_split\n", + "\n", + "X_train, X_valid, Y_train, Y_valid = train_test_split(X, Y, test_size =0.3, random_state=1212) ## 一般如果測試資料集超過1000筆就可以了，所以比率不會設這麼高\n", + "print(X_train.shape) ## (445, 17)\n", + "print(X_valid.shape) ## (446, 17) \n", + "print(Y_train.shape) ## (445,)\n", + "print(Y_valid.shape) ## (446,)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "SVM: 0.6455223880597015\n", + "DecisionTree: 0.7611940298507462\n", + "RandomForest: 0.8544776119402985\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "c:\\users\\user\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\sklearn\\svm\\base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n", + " \"avoid this warning.\", FutureWarning)\n", + "c:\\users\\user\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\sklearn\\ensemble\\forest.py:248: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n", + " \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "AdaBoost: 0.7910447761194029\n", + "XGB: 0.8432835820895522\n" + ] + } + ], + "source": [ + "def get_accuracy(clf):\n", + " #=============your works starts===============#\n", + " clf = clf()\n", + " clf = clf.fit(X_train, Y_train)\n", + " y_pred = clf.predict(X_valid)\n", + " accuracy = (str(sum(Y_valid == y_pred)/Y_valid.shape[0]))\n", + " #==============your works ends================#\n", + " return accuracy\n", + "\n", + "print('SVM: ', get_accuracy(SVC))\n", + "print('DecisionTree: ', get_accuracy(DecisionTreeClassifier))\n", + "print('RandomForest: ', get_accuracy(RandomForestClassifier))\n", + "print('AdaBoost: ', get_accuracy(AdaBoostClassifier)) ## Boosting的演算法\n", + "print('XGB: ', get_accuracy(XGBClassifier))\n", + "\n", + "# SVM: 0.609865470852\n", + "# DecisionTree: 0.764573991031\n", + "# RandomForest: 0.795964125561\n", + "# AdaBoost: 0.784753363229\n", + "# XGB: 0.80269058296" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:18:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "Accuracy: 0.835820895522388\n" + ] + } + ], + "source": [ + "# Set our parameters for xgboost\n", + "params = {}\n", + "# 請填入以下參數: \n", + "# 目標函數: 二元分類\n", + "# 評價函數: logloss\n", + "# 學習速度: 0.04\n", + "# 最大深度: 5\n", + "#=============your works starts===============#\n", + "params['objective'] = 'binary:logistic'\n", + "params['eval_metric'] = 'logloss'\n", + "params['eta'] = 0.04\n", + "params['max_depth'] = 3\n", + "#==============your works ends================#\n", + "\n", + "d_train = xgboost.DMatrix(X_train, label=Y_train)\n", + "d_valid = xgboost.DMatrix(X_valid, label=Y_valid)\n", + "\n", + "watchlist = [(d_train, 'train'), (d_valid, 'valid')]\n", + "\n", + "bst = xgboost.train(params, d_train, 100, watchlist, early_stopping_rounds=100, verbose_eval=0)\n", + "y_pred = bst.predict(xgboost.DMatrix(X_valid))\n", + "print(\"Accuracy: \", str(sum(Y_valid == (y_pred > 0.5))/Y_valid.shape[0]))\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": true + }, + "source": [ + "# 房價資料集" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [], + "source": [ + "import urllib.request\n", + "if 'df_realestate_processed.csv' not in os.listdir():\n", + " url = 'https://s3.amazonaws.com/datasets-jeremy/df_realestate_processed.csv' \n", + " urllib.request.urlretrieve(url, 'df_realestate_processed.csv')\n", + " \n", + "# processed\n", + "path = \"df_realestate_processed.csv\"\n", + "df_realestate_processed = pd.read_csv(path)\n", + "X = df_realestate_processed.drop([\"price_per_meter\", \"total_price\"], axis=1)\n", + "Y = df_realestate_processed['total_price']\n" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [], + "source": [ + "X_train = X.iloc[:-1000]\n", + "Y_train = Y.iloc[:-1000]\n", + "Y_train = np.log(Y_train) \n", + "\n", + "X_valid = X.iloc[-1000:]\n", + "Y_valid = Y.iloc[-1000:]\n", + "Y_valid = np.log(Y_valid) " + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:18:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[0]\ttrain-rmse:15.8112\tvalid-rmse:15.6215\n", + "Multiple eval metrics have been passed: 'valid-rmse' will be used for early stopping.\n", + "\n", + "Will train until valid-rmse hasn't improved in 10 rounds.\n", + "[19:18:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[10]\ttrain-rmse:11.6632\tvalid-rmse:11.5317\n", + "[19:18:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[20]\ttrain-rmse:8.60487\tvalid-rmse:8.51713\n", + "[19:18:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[30]\ttrain-rmse:6.35023\tvalid-rmse:6.30023\n", + "[19:18:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:18:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[40]\ttrain-rmse:4.68857\tvalid-rmse:4.66973\n", + "[19:19:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[50]\ttrain-rmse:3.46457\tvalid-rmse:3.46551\n", + "[19:19:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:19:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[60]\ttrain-rmse:2.56373\tvalid-rmse:2.58316\n", + "[19:19:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[70]\ttrain-rmse:1.90185\tvalid-rmse:1.93611\n", + "[19:19:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[80]\ttrain-rmse:1.41686\tvalid-rmse:1.46651\n", + "[19:19:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[90]\ttrain-rmse:1.06302\tvalid-rmse:1.12422\n", + "[19:19:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[100]\ttrain-rmse:0.806951\tvalid-rmse:0.876425\n", + "[19:19:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:19:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[110]\ttrain-rmse:0.624079\tvalid-rmse:0.700146\n", + "[19:19:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[120]\ttrain-rmse:0.496058\tvalid-rmse:0.577805\n", + "[19:19:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:19:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[130]\ttrain-rmse:0.408926\tvalid-rmse:0.49603\n", + "[19:20:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[140]\ttrain-rmse:0.351423\tvalid-rmse:0.442858\n", + "[19:20:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[150]\ttrain-rmse:0.314738\tvalid-rmse:0.409589\n", + "[19:20:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:20:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[160]\ttrain-rmse:0.291941\tvalid-rmse:0.387703\n", + "[19:20:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[170]\ttrain-rmse:0.277808\tvalid-rmse:0.373504\n", + "[19:20:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[180]\ttrain-rmse:0.26903\tvalid-rmse:0.364483\n", + "[19:20:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[190]\ttrain-rmse:0.263451\tvalid-rmse:0.358139\n", + "[19:20:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:20:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[200]\ttrain-rmse:0.259615\tvalid-rmse:0.352701\n", + "[19:21:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[210]\ttrain-rmse:0.256954\tvalid-rmse:0.349393\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:21:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[220]\ttrain-rmse:0.254944\tvalid-rmse:0.347053\n", + "[19:21:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[230]\ttrain-rmse:0.253445\tvalid-rmse:0.345315\n", + "[19:21:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[240]\ttrain-rmse:0.252139\tvalid-rmse:0.343419\n", + "[19:21:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[250]\ttrain-rmse:0.251006\tvalid-rmse:0.342147\n", + "[19:21:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[260]\ttrain-rmse:0.250072\tvalid-rmse:0.341061\n", + "[19:21:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:21:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:22:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[270]\ttrain-rmse:0.249214\tvalid-rmse:0.340131\n", + "[19:22:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[280]\ttrain-rmse:0.248427\tvalid-rmse:0.339297\n", + "[19:22:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[290]\ttrain-rmse:0.247735\tvalid-rmse:0.338445\n", + "[19:22:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[300]\ttrain-rmse:0.247081\tvalid-rmse:0.33738\n", + "[19:22:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[310]\ttrain-rmse:0.24648\tvalid-rmse:0.337037\n", + "[19:22:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:22:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:23:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[320]\ttrain-rmse:0.245882\tvalid-rmse:0.336335\n", + "[19:23:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[330]\ttrain-rmse:0.245386\tvalid-rmse:0.335948\n", + "[19:23:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[340]\ttrain-rmse:0.244862\tvalid-rmse:0.335533\n", + "[19:23:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[350]\ttrain-rmse:0.244334\tvalid-rmse:0.33528\n", + "[19:23:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[360]\ttrain-rmse:0.243893\tvalid-rmse:0.334766\n", + "[19:23:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:23:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[370]\ttrain-rmse:0.243417\tvalid-rmse:0.334686\n", + "[19:23:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:23:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[380]\ttrain-rmse:0.242968\tvalid-rmse:0.334314\n", + "[19:24:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[390]\ttrain-rmse:0.242505\tvalid-rmse:0.333969\n", + "[19:24:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[400]\ttrain-rmse:0.242057\tvalid-rmse:0.333667\n", + "[19:24:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[410]\ttrain-rmse:0.241676\tvalid-rmse:0.333362\n", + "[19:24:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[420]\ttrain-rmse:0.241232\tvalid-rmse:0.332969\n", + "[19:24:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:24:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:44] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[430]\ttrain-rmse:0.240818\tvalid-rmse:0.332734\n", + "[19:24:45] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:46] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:47] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:48] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:49] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:50] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:51] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[440]\ttrain-rmse:0.24045\tvalid-rmse:0.332447\n", + "[19:24:52] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:53] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:54] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:55] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:56] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:57] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:58] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:24:59] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[450]\ttrain-rmse:0.240066\tvalid-rmse:0.332321\n", + "[19:25:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:00] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:01] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:02] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:03] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:04] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:05] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:06] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[460]\ttrain-rmse:0.239724\tvalid-rmse:0.331935\n", + "[19:25:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:07] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:08] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:09] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:10] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:11] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:12] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:13] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[470]\ttrain-rmse:0.239354\tvalid-rmse:0.331688\n", + "[19:25:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:14] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:15] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:16] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[19:25:17] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:18] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:19] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:20] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[480]\ttrain-rmse:0.238993\tvalid-rmse:0.331617\n", + "[19:25:21] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:22] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:23] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:24] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:25] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:26] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:27] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[490]\ttrain-rmse:0.23864\tvalid-rmse:0.331323\n", + "[19:25:28] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:29] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:30] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:31] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:32] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:33] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:34] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[500]\ttrain-rmse:0.238285\tvalid-rmse:0.331034\n", + "[19:25:35] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:36] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:37] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:38] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:39] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:40] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:41] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[510]\ttrain-rmse:0.23793\tvalid-rmse:0.331124\n", + "[19:25:42] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "[19:25:43] C:\\Users\\Administrator\\Desktop\\xgboost\\src\\tree\\updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 0 pruned nodes, max_depth=3\n", + "Stopping. Best iteration:\n", + "[502]\ttrain-rmse:0.238198\tvalid-rmse:0.33097\n", + "\n" + ] + } + ], + "source": [ + "# Set our parameters for xgboost\n", + "params = {}\n", + "\n", + "# 請填入以下參數: \n", + "# 目標函數: 線性回歸\n", + "# 評價函數: rmse\n", + "# 學習速度: 0.01\n", + "# 最大深度: 5\n", + "# bst = xgboost.train(params, d_train, 3000, watchlist, early_stopping_rounds=50, verbose_eval=10)\n", + "#=============your works starts===============#\n", + "params['objective'] = 'reg:linear'\n", + "params['eval_metric'] = 'rmse'\n", + "params['eta'] = 0.03\n", + "params['max_depth'] = 3\n", + "d_train = xgboost.DMatrix(X_train, label=Y_train)\n", + "d_valid = xgboost.DMatrix(X_valid, label=Y_valid)\n", + "watchlist = [(d_train, 'train'), (d_valid, 'valid')]\n", + "bst = xgboost.train(params, d_train, 3000, watchlist, early_stopping_rounds=10, verbose_eval=10)\n", + "Y_pred = bst.predict(xgboost.DMatrix(X_valid))\n", + "#==============your works ends================#\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#　模型save與load的方式自己看\n", + "# bst.save_model(\"bst_subtotal_log_with_cross.pickle.dat\")\n", + "# bst = xgboost.Booster({'nthread':1}) #init model\n", + "# bst.load_model(\"bst_subtotal_log_with_cross.pickle.dat\") # load data\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# 請使用 xgboost.plot_importance，並設定max_num_features=10\n", + "#!=============your works starts===============!#\n", + "xgboost.plot_importance(bst, max_num_features=10)\n", + "#!==============your works ends================!#\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
predicttrutherror
015413886.015880006.780.029353
112065383.010999982.000.096855
229951496.028199982.040.062110
323521218.021920043.690.073046
45400714.53220663.360.676895
\n", + "
" + ], + "text/plain": [ + " predict truth error\n", + "0 15413886.0 15880006.78 0.029353\n", + "1 12065383.0 10999982.00 0.096855\n", + "2 29951496.0 28199982.04 0.062110\n", + "3 23521218.0 21920043.69 0.073046\n", + "4 5400714.5 3220663.36 0.676895" + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df_result = pd.DataFrame()\n", + "\n", + "# 1. 使用X_valid去評價此模型\n", + "# 2. 使用 ['predict', 'truth', 'error'] 三個欄位的DataFrame去使決畫呈現預測結果\n", + "# (1). 請注意與測結果(Y_pred)與真實值(Y_valid)都必須取exp方能反映實際情況\n", + "# (2). error 請使用計算(predict-truth)/truth計算誤差百分比\n", + "#!=============your works starts===============!#\n", + "Y_pred = bst.predict(xgboost.DMatrix(X_valid))\n", + "df_result['predict'] = np.exp(Y_pred)\n", + "df_result['truth'] = np.exp((list(Y_valid)))\n", + "df_result['error'] = df_result.apply(lambda x:np.abs(x['predict'] - x['truth']) / x['truth'], axis=1)\n", + "df_result_sort = df_result.sort_values('truth')\n", + "#!==============your works ends================!#\n", + "\n", + "df_result.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# 請使用df_result_sort濾掉error大於1的部分畫出error的分布圖\n", + "#!=============your works starts===============!#\n", + "df_result_sort.loc[df_result_sort['error'] < 1, 'error'].plot('hist')\n", + "#!==============your works ends================!#\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# 請使用plt.scatter，以0~len(df_result)作為x，預測值(黑色)與實際值(紅色)作為y。\n", + "#!=============your works starts===============!#\n", + "plt.scatter(range(len(df_result)), df_result_sort['predict'].values, color='black', s=0.5)\n", + "plt.scatter(range(len(df_result)), df_result_sort['truth'].values, color='red', s=0.5)\n", + "#!==============your works ends================!#\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.6.4" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/07_RealEstatePractice/main.ipynb b/07_RealEstatePractice/main07.ipynb similarity index 100% rename from 07_RealEstatePractice/main.ipynb rename to 07_RealEstatePractice/main07.ipynb diff --git a/09_IrImplementation/ProcessedData.json b/08_IrTheory/ProcessedData.json similarity index 100% rename from 09_IrImplementation/ProcessedData.json rename to 08_IrTheory/ProcessedData.json diff --git a/08_IrTheory/cloud_mask7.png b/08_IrTheory/cloud_mask7.png new file mode 100644 index 0000000..2db7ec2 Binary files /dev/null and b/08_IrTheory/cloud_mask7.png differ diff --git a/08_IrTheory/day22_IRTheory.ipynb b/08_IrTheory/day22_IRTheory.ipynb deleted file mode 100644 index 5571b74..0000000 --- a/08_IrTheory/day22_IRTheory.ipynb +++ /dev/null @@ -1,498 +0,0 @@ -{ - "cells": [ - { - "cell_type": "code", - "execution_count": 4, - "metadata": { - "collapsed": true - }, - "outputs": [], - "source": [ - "import pandas as pd\n", - "\n", - "import nltk\n", - "\n", - "from nltk.stem.porter import PorterStemmer\n", - "porter_stemmer = PorterStemmer()\n", - "\n", - "from nltk.corpus import stopwords\n", - "stops = stopwords.words('english')\n", - "from string import punctuation" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
sentence
0what time is it?
1how long has it been since we started?
2that's a long time ago
\n", - "
" - ], - "text/plain": [ - " sentence\n", - "0 what time is it?\n", - "1 how long has it been since we started?\n", - "2 that's a long time ago" - ] - }, - "execution_count": 3, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "corpus = [\"what time is it?\", \"how long has it been since we started?\", \"that's a long time ago\"]\n", - "df = pd.DataFrame(corpus, columns=['sentence'])\n", - "df" - ] - }, - { - "cell_type": "code", - "execution_count": 9, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
sentencetokenize
0what time is it?[what, time, is, it, ?]
1how long has it been since we started?[how, long, has, it, been, since, we, started, ?]
2that's a long time ago[that, 's, a, long, time, ago]
\n", - "
" - ], - "text/plain": [ - " sentence \\\n", - "0 what time is it? \n", - "1 how long has it been since we started? \n", - "2 that's a long time ago \n", - "\n", - " tokenize \n", - "0 [what, time, is, it, ?] \n", - "1 [how, long, has, it, been, since, we, started, ?] \n", - "2 [that, 's, a, long, time, ago] " - ] - }, - "execution_count": 9, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df['tokenize'] = df['sentence'].apply(nltk.word_tokenize)\n", - "df" - ] - }, - { - "cell_type": "code", - "execution_count": 12, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "['what', 'time', 'is', 'it', '?', 'how', 'long', 'has', 'it', 'been', 'since', 'we', 'started', '?', 'that', \"'s\", 'a', 'long', 'time', 'ago']\n" - ] - } - ], - "source": [ - "word_index = []\n", - "for tokens in df['tokenize']:\n", - " word_index.extend(tokens)\n", - "print(word_index)" - ] - }, - { - "cell_type": "code", - "execution_count": 13, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
sentencetokenizewhattimeisit?howlonghasbeensincewestartedthat'saago
0what time is it?[what, time, is, it, ?]0000000000000000
1how long has it been since we started?[how, long, has, it, been, since, we, started, ?]0000000000000000
2that's a long time ago[that, 's, a, long, time, ago]0000000000000000
\n", - "
" - ], - "text/plain": [ - " sentence \\\n", - "0 what time is it? \n", - "1 how long has it been since we started? \n", - "2 that's a long time ago \n", - "\n", - " tokenize what time is it ? \\\n", - "0 [what, time, is, it, ?] 0 0 0 0 0 \n", - "1 [how, long, has, it, been, since, we, started, ?] 0 0 0 0 0 \n", - "2 [that, 's, a, long, time, ago] 0 0 0 0 0 \n", - "\n", - " how long has been since we started that 's a ago \n", - "0 0 0 0 0 0 0 0 0 0 0 0 \n", - "1 0 0 0 0 0 0 0 0 0 0 0 \n", - "2 0 0 0 0 0 0 0 0 0 0 0 " - ] - }, - "execution_count": 13, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "for column in word_index:\n", - " df[column] = 0\n", - "df" - ] - }, - { - "cell_type": "code", - "execution_count": 18, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
sentencetokenizewhattimeisit?howlonghasbeensincewestartedthat'saago
0what time is it?[what, time, is, it, ?]2222200000000000
1how long has it been since we started?[how, long, has, it, been, since, we, started, ?]0002222222220000
2that's a long time ago[that, 's, a, long, time, ago]0200002000002222
\n", - "
" - ], - "text/plain": [ - " sentence \\\n", - "0 what time is it? \n", - "1 how long has it been since we started? \n", - "2 that's a long time ago \n", - "\n", - " tokenize what time is it ? \\\n", - "0 [what, time, is, it, ?] 2 2 2 2 2 \n", - "1 [how, long, has, it, been, since, we, started, ?] 0 0 0 2 2 \n", - "2 [that, 's, a, long, time, ago] 0 2 0 0 0 \n", - "\n", - " how long has been since we started that 's a ago \n", - "0 0 0 0 0 0 0 0 0 0 0 0 \n", - "1 2 2 2 2 2 2 2 0 0 0 0 \n", - "2 0 2 0 0 0 0 0 2 2 2 2 " - ] - }, - "execution_count": 18, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "def build_word_index(row):\n", - " tokens = row['tokenize']\n", - " for token in tokens:\n", - " row[token] += 1\n", - " return row\n", - "df = df.apply(build_word_index, axis=1)\n", - "df" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "collapsed": true - }, - "outputs": [], - "source": [] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.6.4" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/09_IrImplementation/dict.txt.big b/08_IrTheory/dict.txt.big similarity index 100% rename from 09_IrImplementation/dict.txt.big rename to 08_IrTheory/dict.txt.big diff --git a/08_IrTheory/main08.ipynb b/08_IrTheory/main08.ipynb new file mode 100644 index 0000000..71ebff4 --- /dev/null +++ b/08_IrTheory/main08.ipynb @@ -0,0 +1,2507 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 英文NLP基礎教學" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "\n", + "import nltk\n", + "\n", + "from nltk.stem.porter import PorterStemmer\n", + "porter_stemmer = PorterStemmer()\n", + "from nltk.stem.lancaster import LancasterStemmer\n", + "lancaster_stemmer = LancasterStemmer()\n", + "from nltk.stem import SnowballStemmer\n", + "snowball_stemmer = SnowballStemmer('english')\n", + "from nltk.stem import WordNetLemmatizer\n", + "wordnet_lemmatizer = WordNetLemmatizer()\n", + "\n", + "\n", + "from nltk.corpus import stopwords\n", + "stops = stopwords.words('english')\n", + "from string import punctuation" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "This/value/is/also/called/cut-off/in/the/literature/./If/float/,/the/parameter/represents/a/proportion/of/documents/,/integer/absolute/counts/.\n", + "This/value/is/also/called/cut/-/off/in/the/literature/./If/float/,/the/parameter/represents/a/proportion/of/documents/,/integer/absolute/counts/.\n" + ] + } + ], + "source": [ + "testStr = \"This value is also called cut-off in the literature. If float, the parameter represents a proportion of documents, integer absolute counts.\"\n", + "# 請使用nltk.word_tokenize及nltk.wordpunct_tokenize進行分詞，並比較其中差異。\n", + "#=============your works starts===============#\n", + "word_tokenize_tokens = nltk.word_tokenize(testStr)\n", + "wordpunct_tokenize_tokens = nltk.wordpunct_tokenize(testStr)\n", + "#==============your works ends================#\n", + "\n", + "print(\"/\".join(word_tokenize_tokens))\n", + "print(\"/\".join(wordpunct_tokenize_tokens))\n", + "# This/value/is/also/called/cut-off/in/the/literature/./If/float/,/the/parameter/represents/a/proportion/of/documents/,/integer/absolute/counts/.\n", + "# This/value/is/also/called/cut/-/off/in/the/literature/./If/float/,/the/parameter/represents/a/proportion/of/documents/,/integer/absolute/counts/." + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "['thi', 'thi', 'this', 'This']\n", + "['valu', 'valu', 'valu', 'value']\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
porter_stemmerlancaster_stemmersnowball_stemmerwordnet_lemmatizer
ThisthithithisThis
valuevaluvaluvaluvalue
isisisisis
alsoalsoalsoalsoalso
calledcallcalcallcalled
cutcutcutcutcut
-----
offoffoffoffoff
ininininin
thethethethethe
literatureliteraturlitliteraturliterature
.....
IfIfififIf
floatfloatflofloatfloat
,,,,,
thethethethethe
parameterparametparametparametparameter
representsrepresrepresrepresrepresents
aaaaa
proportionproportproportproportproportion
ofofofofof
documentsdocumentdocudocumentdocument
,,,,,
integerinteginteginteginteger
absoluteabsolutabsolvabsolutabsolute
countscountcountcountcount
.....
\n", + "
" + ], + "text/plain": [ + " porter_stemmer lancaster_stemmer snowball_stemmer \\\n", + "This thi thi this \n", + "value valu valu valu \n", + "is is is is \n", + "also also also also \n", + "called call cal call \n", + "cut cut cut cut \n", + "- - - - \n", + "off off off off \n", + "in in in in \n", + "the the the the \n", + "literature literatur lit literatur \n", + ". . . . \n", + "If If if if \n", + "float float flo float \n", + ", , , , \n", + "the the the the \n", + "parameter paramet paramet paramet \n", + "represents repres repres repres \n", + "a a a a \n", + "proportion proport proport proport \n", + "of of of of \n", + "documents document docu document \n", + ", , , , \n", + "integer integ integ integ \n", + "absolute absolut absolv absolut \n", + "counts count count count \n", + ". . . . \n", + "\n", + " wordnet_lemmatizer \n", + "This This \n", + "value value \n", + "is is \n", + "also also \n", + "called called \n", + "cut cut \n", + "- - \n", + "off off \n", + "in in \n", + "the the \n", + "literature literature \n", + ". . \n", + "If If \n", + "float float \n", + ", , \n", + "the the \n", + "parameter parameter \n", + "represents represents \n", + "a a \n", + "proportion proportion \n", + "of of \n", + "documents document \n", + ", , \n", + "integer integer \n", + "absolute absolute \n", + "counts count \n", + ". . " + ] + }, + "execution_count": 18, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "tokens = wordpunct_tokenize_tokens\n", + "df = pd.DataFrame(index = tokens)\n", + "# 請使用porter_stemmer, lancaster_stemmer, snowball_stemmer, wordnet_lemmatizer，進行stemming或是lemmatize，並放到一個df比較其中差異\n", + "#=============your works starts===============#\n", + "df['porter_stemmer'] = [porter_stemmer.stem(t) for t in tokens]\n", + "df['lancaster_stemmer'] = [lancaster_stemmer.stem(t) for t in tokens]\n", + "df['snowball_stemmer'] = [snowball_stemmer.stem(t) for t in tokens]\n", + "df['wordnet_lemmatizer'] = [wordnet_lemmatizer.lemmatize(t) for t in tokens]\n", + "#==============your works ends================#\n", + "\n", + "print(df.iloc[0].tolist())\n", + "print(df.iloc[1].tolist())\n", + "# ['thi', 'thi', 'this', 'This']\n", + "# ['valu', 'valu', 'valu', 'value']\n", + "\n", + "df" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "標點符號\n", + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~\n", + "停用字\n", + "['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', \"you're\", \"you've\", \"you'll\", \"you'd\", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', 'she', \"she's\", 'her', 'hers', 'herself', 'it', \"it's\", 'its', 'itself', 'they', 'them', 'their', 'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', \"that'll\", 'these', 'those', 'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', 'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', 'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after', 'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further', 'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more', 'most', 'other', 'some', 'such', 'no', 'nor', 'not', 'only', 'own', 'same', 'so', 'than', 'too', 'very', 's', 't', 'can', 'will', 'just', 'don', \"don't\", 'should', \"should've\", 'now', 'd', 'll', 'm', 'o', 're', 've', 'y', 'ain', 'aren', \"aren't\", 'couldn', \"couldn't\", 'didn', \"didn't\", 'doesn', \"doesn't\", 'hadn', \"hadn't\", 'hasn', \"hasn't\", 'haven', \"haven't\", 'isn', \"isn't\", 'ma', 'mightn', \"mightn't\", 'mustn', \"mustn't\", 'needn', \"needn't\", 'shan', \"shan't\", 'shouldn', \"shouldn't\", 'wasn', \"wasn't\", 'weren', \"weren't\", 'won', \"won't\", 'wouldn', \"wouldn't\"]\n" + ] + } + ], + "source": [ + "print(\"標點符號\")\n", + "print(punctuation)\n", + "print(\"停用字\")\n", + "print(stops)" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
porter_stemmerlancaster_stemmersnowball_stemmerwordnet_lemmatizer
Thisthithithisthis
valuevaluvaluvaluvalue
alsoalsoalsoalsoalso
calledcallcalcallcalled
cutcutcutcutcut
literatureliteraturlitliteraturliterature
Ififififif
floatfloatflofloatfloat
parameterparametparametparametparameter
representsrepresrepresrepresrepresents
proportionproportproportproportproportion
documentsdocumentdocudocumentdocument
integerinteginteginteginteger
absoluteabsolutabsolvabsolutabsolute
countscountcountcountcount
\n", + "
" + ], + "text/plain": [ + " porter_stemmer lancaster_stemmer snowball_stemmer \\\n", + "This thi thi this \n", + "value valu valu valu \n", + "also also also also \n", + "called call cal call \n", + "cut cut cut cut \n", + "literature literatur lit literatur \n", + "If if if if \n", + "float float flo float \n", + "parameter paramet paramet paramet \n", + "represents repres repres repres \n", + "proportion proport proport proport \n", + "documents document docu document \n", + "integer integ integ integ \n", + "absolute absolut absolv absolut \n", + "counts count count count \n", + "\n", + " wordnet_lemmatizer \n", + "This this \n", + "value value \n", + "also also \n", + "called called \n", + "cut cut \n", + "literature literature \n", + "If if \n", + "float float \n", + "parameter parameter \n", + "represents represents \n", + "proportion proportion \n", + "documents document \n", + "integer integer \n", + "absolute absolute \n", + "counts count " + ] + }, + "execution_count": 20, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df = pd.DataFrame(index = [t for t in tokens if t not in stops and t not in punctuation])\n", + "# 請使用porter_stemmer, lancaster_stemmer, snowball_stemmer, wordnet_lemmatizer，進行stemming或是lemmatize，並放到一個df比較其中差異\n", + "# 請去除標點符號與停用字\n", + "#=============your works starts===============#\n", + "df['porter_stemmer'] = [porter_stemmer.stem(t.lower()) for t in tokens if t not in stops and t not in punctuation]\n", + "df['lancaster_stemmer'] = [lancaster_stemmer.stem(t.lower()) for t in tokens if t not in stops and t not in punctuation]\n", + "df['snowball_stemmer'] = [snowball_stemmer.stem(t.lower()) for t in tokens if t not in stops and t not in punctuation]\n", + "df['wordnet_lemmatizer'] = [wordnet_lemmatizer.lemmatize(t.lower()) for t in tokens if t not in stops and t not in punctuation]\n", + "#==============your works ends================#\n", + "\n", + "df" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
defaultuniversal
ThisDTDET
valueNNNOUN
isVBZVERB
alsoRBADV
calledVBNVERB
cutVBNVERB
-:.
offRBADV
inINADP
theDTDET
literatureNNNOUN
...
IfINADP
floatNNNOUN
,,.
theDTDET
parameterNNNOUN
representsVBZVERB
aDTDET
proportionNNNOUN
ofINADP
documentsNNSNOUN
,,.
integerNNNOUN
absoluteNNNOUN
countsNNSNOUN
...
\n", + "
" + ], + "text/plain": [ + " default universal\n", + "This DT DET\n", + "value NN NOUN\n", + "is VBZ VERB\n", + "also RB ADV\n", + "called VBN VERB\n", + "cut VBN VERB\n", + "- : .\n", + "off RB ADV\n", + "in IN ADP\n", + "the DT DET\n", + "literature NN NOUN\n", + ". . .\n", + "If IN ADP\n", + "float NN NOUN\n", + ", , .\n", + "the DT DET\n", + "parameter NN NOUN\n", + "represents VBZ VERB\n", + "a DT DET\n", + "proportion NN NOUN\n", + "of IN ADP\n", + "documents NNS NOUN\n", + ", , .\n", + "integer NN NOUN\n", + "absolute NN NOUN\n", + "counts NNS NOUN\n", + ". . ." + ] + }, + "execution_count": 22, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df_tag = pd.DataFrame(index = tokens)\n", + "# 請使用nltk.pos_tag進行詞性標記，並嘗試設定參數tagset='universal'\n", + "#=============your works starts===============#\n", + "df_tag['default'] = [tag for term, tag in nltk.pos_tag(tokens)]\n", + "df_tag['universal'] = [tag for term, tag in nltk.pos_tag(tokens, tagset='universal')]\n", + "#==============your works ends================#\n", + "\n", + "df_tag" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 建立詞向量" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import pandas as pd\n", + "import nltk\n", + "from nltk.stem.porter import PorterStemmer\n", + "porter_stemmer = PorterStemmer()\n", + "\n", + "from nltk.corpus import stopwords\n", + "stops = stopwords.words('english')\n", + "from string import punctuation" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
sentence
0what time is it?
1how long has it been since we started?
2that's a long time ago
\n", + "
" + ], + "text/plain": [ + " sentence\n", + "0 what time is it?\n", + "1 how long has it been since we started?\n", + "2 that's a long time ago" + ] + }, + "execution_count": 24, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "corpus = [\"what time is it?\", \"how long has it been since we started?\", \"that's a long time ago\"]\n", + "df = pd.DataFrame(corpus, columns=['sentence'])\n", + "df" + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "0 [what, time, is, it, ?]\n", + "1 [how, long, has, it, been, since, we, started, ?]\n", + "2 [that, 's, a, long, time, ago]\n", + "Name: tokenize, dtype: object" + ] + }, + "execution_count": 25, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# 請使用nltk.word_tokenize將每一行的詞彙切開\n", + "#=============your works starts===============#\n", + "df['tokenize'] = df['sentence'].apply(nltk.word_tokenize)\n", + "#==============your works ends================#\n", + "\n", + "\n", + "df['tokenize']\n", + "# 0 [what, time, is, it, ?]\n", + "# 1 [how, long, has, it, been, since, we, started, ?]\n", + "# 2 [that, 's, a, long, time, ago]\n", + "# Name: tokenize, dtype: object" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "len(word_index) 16\n", + "{'that', 'it', 'long', 'since', 'started', 'ago', 'has', 'a', 'been', 'time', '?', 'is', 'we', 'how', 'what', \"'s\"}\n" + ] + } + ], + "source": [ + "# 請找出不重複的所有出現過的字\n", + "#=============your works starts===============#\n", + "word_index = set(np.hstack([tokens for tokens in df['tokenize']]))\n", + "#==============your works ends================#\n", + "\n", + "print(\"len(word_index)\", len(word_index))\n", + "print(word_index)\n", + "# len(word_index) 16\n", + "# {'is', 'that', 'time', 'long', 'we', 'ago', 'started', 'has', 'been', 'a', \"'s\", 'how', 'what', 'it', 'since', '?'}" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Index(['sentence', 'tokenize', 'that', 'it', 'long', 'since', 'started', 'ago',\n", + " 'has', 'a', 'been', 'time', '?', 'is', 'we', 'how', 'what', ''s'],\n", + " dtype='object')\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
sentencetokenizethatitlongsincestartedagohasabeentime?iswehowwhat's
0what time is it?[what, time, is, it, ?]0000000000000000
1how long has it been since we started?[how, long, has, it, been, since, we, started, ?]0000000000000000
2that's a long time ago[that, 's, a, long, time, ago]0000000000000000
\n", + "
" + ], + "text/plain": [ + " sentence \\\n", + "0 what time is it? \n", + "1 how long has it been since we started? \n", + "2 that's a long time ago \n", + "\n", + " tokenize that it long since \\\n", + "0 [what, time, is, it, ?] 0 0 0 0 \n", + "1 [how, long, has, it, been, since, we, started, ?] 0 0 0 0 \n", + "2 [that, 's, a, long, time, ago] 0 0 0 0 \n", + "\n", + " started ago has a been time ? is we how what 's \n", + "0 0 0 0 0 0 0 0 0 0 0 0 0 \n", + "1 0 0 0 0 0 0 0 0 0 0 0 0 \n", + "2 0 0 0 0 0 0 0 0 0 0 0 0 " + ] + }, + "execution_count": 27, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "for column in word_index:\n", + " # 請幫每一個字創造一個欄位並指派為0\n", + " #=============your works starts===============#\n", + " df[column] = 0\n", + " #==============your works ends================#\n", + "\n", + " \n", + "print(df.columns)\n", + "df\n", + "# Index(['sentence', 'tokenize', 'is', 'that', 'time', 'long', 'we', 'ago',\n", + "# 'started', 'has', 'been', 'a', ''s', 'how', 'what', 'it', 'since', '?'],\n", + "# dtype='object')" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "['what time is it?', ['what', 'time', 'is', 'it', '?'], 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 1, 0]\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
sentencetokenizethatitlongsincestartedagohasabeentime?iswehowwhat's
0what time is it?[what, time, is, it, ?]0100000001110010
1how long has it been since we started?[how, long, has, it, been, since, we, started, ?]0111101010101100
2that's a long time ago[that, 's, a, long, time, ago]1010010101000001
\n", + "
" + ], + "text/plain": [ + " sentence \\\n", + "0 what time is it? \n", + "1 how long has it been since we started? \n", + "2 that's a long time ago \n", + "\n", + " tokenize that it long since \\\n", + "0 [what, time, is, it, ?] 0 1 0 0 \n", + "1 [how, long, has, it, been, since, we, started, ?] 0 1 1 1 \n", + "2 [that, 's, a, long, time, ago] 1 0 1 0 \n", + "\n", + " started ago has a been time ? is we how what 's \n", + "0 0 0 0 0 0 1 1 1 0 0 1 0 \n", + "1 1 0 1 0 1 0 1 0 1 1 0 0 \n", + "2 0 1 0 1 0 1 0 0 0 0 0 1 " + ] + }, + "execution_count": 28, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "def build_word_index(row):\n", + " tokens = row['tokenize']\n", + " for token in tokens:\n", + " # 請幫計算每個字，在這個句子中出現的次數\n", + " #=============your works starts===============#\n", + " row[token] += 1\n", + " #==============your works ends================#\n", + " return row\n", + "\n", + "df_processed = df.apply(build_word_index, axis=1)\n", + "\n", + "print(df_processed.iloc[0].tolist())\n", + "# ['what time is it?', ['what', 'time', 'is', 'it', '?'], 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1]\n", + "df_processed" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 中文NLP教學" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "metadata": {}, + "outputs": [], + "source": [ + "import jieba\n", + "jieba.set_dictionary('dict.txt.big') # 如果是使用繁體文字，請記得去下載繁體字典來使用\n", + "with open('stops.txt', 'r', encoding='utf8') as f: # 中文的停用字，我也忘記從哪裡拿到的，效果還可以，繁體字的資源真的比較少，大家將就一下吧\n", + " stops = f.read().split('\\n') " + ] + }, + { + "cell_type": "code", + "execution_count": 47, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "['下雨天', '留客', '天留', '我', '不留']\n", + "['下雨', '下雨天', '雨天', '留客', '天', '留', '我', '不留']\n" + ] + } + ], + "source": [ + "# 請使用jieba.cut進行斷詞，並嘗試使用全斷詞模式(cut_all=True)\n", + "#=============your works starts===============#\n", + "result_cut = [t for t in jieba.cut('下雨天留客天留我不留')]\n", + "result_cutall = [t for t in jieba.cut('下雨天留客天留我不留', cut_all=True)]\n", + "#==============your works ends================#\n", + "\n", + "\n", + "print(result_cut)\n", + "print(result_cutall)\n", + "# ['下雨天', '留客', '天留', '我', '不留']\n", + "# ['下雨', '下雨天', '雨天', '留客', '天', '留', '我', '不留']" + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "停用字\n", + "['\\ufeff\\ufeff,', '', '、', '。', '“', '”', '《', '》', '！', '，', '：', '；', '？', '人民', '末##末', '啊', '阿', '哎', '哎呀', '哎喲', '唉', '我', '我們', '按', '按照', '依照', '吧', '吧噠', '把', '罷了', '被', '本', '本著', '比', '比方', '比如', '鄙人', '彼', '彼此', '邊', '別', '別的', '別說', '並', '並且', '不比', '不成', '不單', '不但', '不獨', '不管', '不光', '不過', '不僅', '不拘', '不論', '不怕', '不然', '不如', '不特', '不惟', '不問', '不只', '朝', '朝著', '趁', '趁著', '乘', '沖', '除', '除此之外', '除非', '除了', '此', '此間', '此外', '從', '從而', '打', '待', '但', '但是', '當', '當著', '到', '得', '的', '的話', '等', '等等', '地', '第', '叮咚', '對', '對於', '多', '多少', '而', '而況', '而且']\n" + ] + } + ], + "source": [ + "print(\"停用字\")\n", + "print(stops[:100])" + ] + }, + { + "cell_type": "code", + "execution_count": 48, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[('語言', 10), ('學習', 10), ('機器', 9), ('研究', 8), ('系統', 8), ('資料', 8), ('翻譯', 7), ('模型', 7), ('處理', 6), ('年代', 6), ('機器翻譯', 5), ('自然', 4), ('自然語言', 4), ('年', 4), ('統計', 4), ('演算', 4), ('演算法', 4), ('算法', 4), ('早期', 3), ('問題', 3), ('預期', 3), ('一直', 3), ('直到', 3), ('1980', 3), ('發展', 3), ('成功', 3), ('NLP', 3), ('一個', 3), ('世界', 3), ('設計', 3), ('許多', 3), ('人工', 3), ('規則', 3), ('基礎', 3), ('語料', 3), ('語料庫', 3), ('輸入', 3), ('這種', 3), ('監督', 3), ('1950', 2), ('作', 2), ('圖靈', 2), ('電腦', 2), ('智慧', 2), ('智慧型', 2), ('提出', 2), ('現在', 2), ('1966', 2), ('未', 2), ('末期', 2), ('展出', 2), ('得以', 2), ('特別', 2), ('包括', 2), ('一種', 2), ('ELIZA', 2), ('類似', 2), ('回答', 2), ('頭痛', 2), ('程式', 2), ('資訊', 2), ('架構', 2), ('能夠', 2), ('複雜', 2), ('訂定', 2), ('產生', 2), ('理論', 2), ('聚焦', 2), ('裡', 2), ('通常', 2), ('非', 2), ('尤其', 2), ('包含', 2), ('督學', 2), ('註解', 2), ('不準', 2), ('準確', 2), ('是從', 1), ('更早', 1), ('發表', 1), ('論文', 1), ('器', 1), ('出現', 1), ('所謂', 1), ('測試', 1), ('判斷', 1), ('條件', 1), ('\\n1954', 1), ('喬治', 1), ('喬治城', 1), ('實驗', 1), ('涉及', 1), ('全部', 1), ('自動', 1), ('超過', 1), ('60', 1), ('句', 1), ('俄文', 1), ('文成', 1), ('英文', 1), ('人員', 1), ('聲稱', 1), ('五年', 1), ('之內', 1), ('即可', 1), ('解決', 1), ('實際', 1), ('進展', 1), ('遠', 1), ('低於', 1), ('ALPAC', 1), ('報告', 1), ('告發', 1), ('發現', 1), ('十年', 1), ('達', 1), ('目標', 1), ('經費', 1), ('遭到', 1), ('大幅', 1), ('削減', 1), ('更上一層', 1), ('更上一層樓', 1), ('一層', 1), ('一層樓', 1), ('層樓', 1), ('\\n1960', 1), ('代發', 1), ('別成', 1), ('統包', 1), ('SHRDLU', 1), ('詞彙', 1), ('設限', 1), ('運作', 1), ('受限', 1), ('積木', 1), ('1964', 1), ('約瑟', 1), ('約瑟夫', 1), ('瑟夫', 1), ('維', 1), ('森', 1), ('鮑', 1), ('姆', 1), ('類比', 1), ('中心', 1), ('治療', 1), ('運用', 1), ('用人', 1), ('人類', 1), ('思想', 1), ('感情', 1), ('訊息', 1), ('有時', 1), ('有時候', 1), ('卻', 1), ('呈現', 1), ('令人', 1), ('訝異', 1), ('異地', 1), ('人', 1), ('之間', 1), ('互動', 1), ('病人', 1), ('超出', 1), ('極小', 1), ('知識', 1), ('範圍', 1), ('之時', 1), ('得到', 1), ('空泛', 1), ('說', 1), ('\\n1970', 1), ('程式設計', 1), ('設計師', 1), ('概念', 1), ('本體', 1), ('本體論', 1), ('conceptual', 1), ('ontologies', 1), ('現實', 1), ('構成', 1), ('成電', 1), ('理解', 1), ('實例', 1), ('MARGIE', 1), ('SAM', 1), ('PAM', 1), ('TaleSpin', 1), ('QUALM', 1), ('Politics', 1), ('Plot', 1), ('Unit', 1), ('聊天', 1), ('天機', 1), ('機器人', 1), ('一時', 1), ('一時期', 1), ('時期', 1), ('寫成', 1), ('PARRY', 1), ('Racter', 1), ('Jabberwacky', 1), ('多數', 1), ('一套', 1), ('引進', 1), ('革新', 1), ('成因', 1), ('兩個', 1), ('運算', 1), ('能力', 1), ('穩定', 1), ('增加', 1), ('參見', 1), ('摩爾', 1), ('摩爾定律', 1), ('定律', 1), ('喬姆斯基', 1), ('姆斯基', 1), ('語言學', 1), ('學理', 1), ('漸漸', 1), ('喪失', 1), ('失主', 1), ('主導', 1), ('轉換', 1), ('生成', 1), ('成文', 1), ('成文法', 1), ('文法', 1), ('傾向', 1), ('語言所', 1), ('所用', 1), ('方法', 1), ('最早', 1), ('使用', 1), ('決策', 1), ('決策樹', 1), ('硬性', 1), ('組成', 1), ('當時', 1), ('既有', 1), ('詞性', 1), ('標記', 1), ('隱', 1), ('馬爾可', 1), ('馬爾可夫', 1), ('可夫', 1), ('引入', 1), ('軟性', 1), ('機率', 1), ('做', 1), ('決定', 1), ('特性', 1), ('賦予', 1), ('代表', 1), ('份量', 1), ('數值', 1), ('語音', 1), ('識別', 1), ('別現', 1), ('現今', 1), ('依賴', 1), ('取', 1), ('即是', 1), ('例子', 1), ('足以', 1), ('有錯', 1), ('錯誤', 1), ('真實', 1), ('真實世界', 1), ('總', 1), ('免不了', 1), ('整合', 1), ('合到', 1), ('多個', 1), ('個子', 1), ('任務', 1), ('較大', 1), ('大系', 1), ('時', 1), ('比較', 1), ('可靠', 1), ('屬於', 1), ('領域', 1), ('歸功', 1), ('IBM', 1), ('漸次', 1), ('以利', 1), ('利用', 1), ('加拿', 1), ('加拿大', 1), ('歐盟', 1), ('現有', 1), ('法律', 1), ('規定', 1), ('政府', 1), ('會議', 1), ('翻譯成', 1), ('譯成', 1), ('所有', 1), ('官方', 1), ('官方語言', 1), ('大部', 1), ('大部分', 1), ('部分', 1), ('分系統', 1), ('打造', 1), ('直到現在', 1), ('限制', 1), ('主要', 1), ('因素', 1), ('大量', 1), ('致力', 1), ('致力於', 1), ('有限', 1), ('有效', 1), ('地學', 1), ('式', 1), ('從沒', 1), ('有人', 1), ('解理', 1), ('理想', 1), ('答案', 1), ('困難', 1), ('同量', 1), ('下', 1), ('料量', 1), ('巨', 1), ('全球', 1), ('資訊網', 1), ('彌補', 1), ('缺點', 1), ('近年', 1), ('年來', 1), ('深度', 1), ('技巧', 1), ('紛紛', 1), ('出爐', 1), ('方面', 1), ('獲得', 1), ('最', 1), ('尖端', 1), ('端的', 1), ('成果', 1), ('語法', 1), ('語法分析', 1), ('法分析', 1), ('分析', 1)]\n" + ] + } + ], + "source": [ + "from collections import Counter\n", + "from wordcloud import WordCloud\n", + "from matplotlib import pyplot as plt\n", + "\n", + "testStr = \"\"\"自然語言處理大體是從1950年代開始，雖然更早期也有作為。1950年，圖靈發表論文「電腦器與智慧型」，提出現在所謂的「圖靈測試」作為判斷智慧型的條件。\n", + "1954年的喬治城實驗涉及全部自動翻譯超過60句俄文成為英文。研究人員聲稱三到五年之內即可解決機器翻譯的問題。[1]不過實際進展遠低於預期，1966年的ALPAC報告發現十年研究未達預期目標，機器翻譯的研究經費遭到大幅削減。一直到1980年代末期，統計機器翻譯系統發展出來，機器翻譯的研究才得以更上一層樓。\n", + "1960年代發展特別成功的NLP系統包括SHRDLU——一個詞彙設限、運作於受限如「積木世界」的一種自然語言系統，以及1964-1966年約瑟夫·維森鮑姆類比「個人中心治療」而設計的ELIZA——幾乎未運用人類思想和感情的訊息，有時候卻能呈現令人訝異地類似人之間的互動。「病人」提出的問題超出ELIZA 極小的知識範圍之時，可能會得到空泛的回答。例如問題是「我的頭痛」，回答是「為什麼說你頭痛？」\n", + "1970年代，程式設計師開始設計「概念本體論」（conceptual ontologies）的程式，將現實世界的資訊，架構成電腦能夠理解的資料。實例有MARGIE、SAM、PAM、TaleSpin、QUALM、Politics以及Plot Unit。許多聊天機器人在這一時期寫成，包括PARRY 、Racter 以及Jabberwacky 。\n", + "一直到1980年代，多數自然語言處理系統是以一套複雜、人工訂定的規則為基礎。不過從1980年代末期開始，語言處理引進了機器學習的演算法，NLP產生革新。成因有兩個：運算能力穩定增加（參見摩爾定律）；以及喬姆斯基 語言學理論漸漸喪失主導（例如轉換-生成文法）。該理論的架構不傾向於語料庫——機器學習處理語言所用方法的基礎。有些最早期使用的機器學習演算法，例如決策樹，是硬性的、「如果-則」規則組成的系統，類似當時既有的人工訂定的規則。不過詞性標記將隱馬爾可夫模型引入NLP，並且研究日益聚焦於軟性的、以機率做決定的統計模型，基礎是將輸入資料裡每一個特性賦予代表其份量的數值。許多語音識別現今依賴的快取語言模型即是一種統計模型的例子。這種模型通常足以處理非預期的輸入資料，尤其是輸入有錯誤（真實世界的資料總免不了），並且在整合到包含多個子任務的較大系統時，結果比較可靠。\n", + "許多早期的成功屬於機器翻譯領域，尤其歸功IBM的研究，漸次發展出更複雜的統計模型。這些系統得以利用加拿大和歐盟現有的語料庫，因為其法律規定政府的會議必須翻譯成所有的官方語言。不過，其他大部分系統必須特別打造自己的語料庫，一直到現在這都是限制其成功的一個主要因素，於是大量的研究致力於從有限的資料更有效地學習。\n", + "近來的研究更加聚焦於非監督式學習和半監督學習的演算法。這種演算法，能夠從沒有人工註解理想答案的資料裡學習。大體而言，這種學習比監督學習困難，並且在同量的資料下，通常產生的結果較不準確。不過沒有註解的資料量極巨（包含了全球資訊網），彌補了較不準確的缺點。\n", + "近年來, 深度學習技巧紛紛出爐[2][3] 在自然語言處理方面獲得最尖端的成果，例如語言模型[4]、語法分析[5][6]等等。\"\"\"\n", + "\n", + "# 請用全斷詞模式對上面文章段落進行分詞\n", + "# 然後計算每一個詞彙出現過的次數，並將出現較多次的排到前面\n", + "#=============your works starts===============#\n", + "terms = [t for t in jieba.cut(testStr, cut_all=True) if t not in stops]\n", + "terms_sorted = sorted(Counter(terms).items(), key=lambda x:x[1], reverse=True)\n", + "#==============your works ends================#\n", + "\n", + "print(terms_sorted)" + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "wordcloud = WordCloud(font_path=\"simsun.ttf\") # 注意必須放數中文字體，否則會變亂碼\n", + "wordcloud.generate_from_frequencies(frequencies=Counter(terms))\n", + "plt.figure(figsize=(15,15))\n", + "plt.imshow(wordcloud, interpolation=\"bilinear\")\n", + "plt.axis(\"off\")\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 46, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "from PIL import Image\n", + "\n", + "alice_mask = np.array(Image.open(\"cloud_mask7.png\"))\n", + "wc = WordCloud(background_color=\"white\", max_words=2000, mask=alice_mask, font_path=\"simsun.ttf\")\n", + "wc.generate_from_frequencies(Counter(terms))\n", + "\n", + "# show\n", + "plt.imshow(wc, interpolation='bilinear')\n", + "plt.axis(\"off\")\n", + "plt.figure()\n", + "plt.imshow(alice_mask, cmap=plt.cm.gray, interpolation='bilinear')\n", + "plt.axis(\"off\")\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": true + }, + "source": [ + "# 中文檢索系統" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "1. TFIDF\n", + "$$TFIDF_{td} = TF_{td} \\times log(\\frac{N}{DF_t})$$\n", + " - 所謂TFIDF應分成兩個部分來理解：TF(Term Frequency)以及IDF(Inverted Document Frequency)。\n", + " - TF(Term Frequency): $TF_{td}$指得是在特定的文章d中特定的字t出現了幾次。這個部分同時，也表示了一個文字在一篇文章的重要性，依但出現越多次，這個字也就越能代表這篇文章。\n", + " - IDF(Inverted Document Frequency): N指得是總共有機篇文章，$DF_t$中的DF是Document Frequency的意思，DFt則是詞彙t在幾篇文章中出現過。$\\frac{DF_t}{N}$也就是所有文章當中，詞彙t在幾篇文章出現過，而其倒數則是Inverted Documnet Index，表著這個詞彙如果在很多文章裏面都出現過，則其重要性會受到懲罰，而取log則只是讓他在分數的影響上比較平滑而已。\n", + " \n", + " \n", + "2. Cosine Similarity\n", + "$$\\cos{\\theta} = \\frac{A \\cdot B}{\\| {A} \\|_2 \\| {B} \\|_2}$$\n", + " - if $A = [1,2,0,4]$ and $B = [3,2,1,0]$\n", + " - $\\cos{\\theta} = \\frac{1 \\cdot 3 + 2 \\cdot 2 + 0 \\cdot 1 + 4 \\cdot 0} {\\sqrt{1^2+2^2+0^2+4^2} \\cdot \\sqrt{3^2+2^2+1^2+0^2}}$" + ] + }, + { + "cell_type": "code", + "execution_count": 54, + "metadata": {}, + "outputs": [], + "source": [ + "import jieba\n", + "jieba.set_dictionary('dict.txt.big') # 如果是使用繁體文字，請記得去下載繁體字典來使用\n", + "import numpy as np\n", + "import pandas as pd" + ] + }, + { + "cell_type": "code", + "execution_count": 55, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
questionans
0小孩出生後應於何時申請育兒津貼?1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在...
1小孩出生後應於何時申請育兒津貼?隨時提出;津貼經審查通過後，追溯自受理申請之當月起發給。兒童出生後六十日內向戶政事務所完成出...
2育兒津貼申請應備文件為何?申請資料應備齊:(一)兒童之戶口名簿影本。(二)申請人之郵局存摺封面影本。(三)父母雙方身分...
3若民眾夫妻雙方均失業，是否可申請家庭育兒津貼費用補助一、育兒津貼補助對象：1.育有二足歲以下兒童。2.兒童之父母至少一方因育兒需要，致未能就業者...
4育兒津貼補助對象為何？育兒津貼補助對象，應符合下列規定：(一)育有二足歲以下兒童。(二)兒童之父母(或監護人)至少...
5如何認定是否符合父母未就業家庭育兒津貼中的未就業？1.未參加勞工「就業」保險（或軍保、公保）。2.最近一年度之稅捐稽徵機關核定之薪資所得及執行...
6未就業家庭育兒津貼條件中，若一方無工作但有投保公會勞保，請問是否可申請該項津貼，請提供該項業...針對勞工保險投保職業工會者（或雇主）等對象，一般均未參加「勞工就業保險」（自行加保者除外），...
7就業者家庭部分托育費用補助所稱的「就業者」為何？(1)受僱於政府、學校或公民營事業單位者。(2)勞動基準法所稱受雇主僱用從事工作者。(3)依...
8我未來想當合格托育人員，但又不是相關科系，新竹市哪裡會開辦托育人員訓練課程呢？勞動部勞動力發展署及本府委託社團法人新竹市社區婦女關懷協會、社團法人新竹市嬰幼兒保育協會、新...
9我該如何成為居家托育服務中心(原社區保母系統)的居家式托育人員呢？欲申請加入居家托育服務中心內的居家托育人員依規需年滿20歲並符合以下資格之一：1.取得保母技...
10親戚朋友托育，是否也可以領補助呢？一般民眾(含寶寶的爺爺奶奶、阿姨、叔叔、舅舅等三親等以內親屬)必須上課126小時托育人員訓練...
11就業者家庭部分托育費用補助申請應備文件為何？向誰申請1.補助申請表。2.最近戶籍謄本或戶口名簿影本。3.郵局存簿封面影本。4.托育契約書（親屬托...
12育兒津貼補助內容為為何？本津貼補助對象，應符合下列規定：(一)育有2足歲以下兒童。(二)兒童之父母(或監護人)至少一...
13誰可以提出育兒津貼申請手續？(一)申請人資格規定如下：1.育有二足歲以下兒童。2.兒童之父母(或監護人)至少一方因育兒需...
14托育費用補助標準為何？補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托...
15托育費用補助對象為何？補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總...
16育兒津貼補助金額多少？育兒津貼補助金額依家庭狀況而定，補助金額如下：(一)低收入戶：每名兒童每月補助新臺幣5000...
17請提供本年度保母系統(居家托育服務中心)相關資訊為何?目前本市居家托育服務中心承辦單位及聯絡電話如下：東區居家托育服務中心承辦單位：新竹市私立光復...
18育兒津貼受理單位為何？審核程序為何？本津貼申領及發放程序規定如下：(一)由申請人檢具相關證明文件向兒童戶籍地之區公所提出申請。(...
19要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服...
20就業者家庭部分托育費用補助申請對象及審核程序為何？就業者家庭部分托育費用補助的實施對象為：1.父母(或監護人)雙方或單親一方皆就業，或父母一方...
21就業者家庭部分托育費用補助審核程序為何？父母需於托育事實發生日起15日內，檢齊應備文件送交幼兒托育地點之社區保母系統或托嬰中心，其補...
22托育補助金額多少，補助期間多長？依各類家庭條件及送托托育人員資格，其托育補助金額說明如下：(一)將嬰幼兒送至托育服務中心中有...
23免費育兒指導的內容是什麼？政府建構友善的托育環境，除了提供托育費用及育兒津貼的經濟補助外，仍然重視親職教育，所以提供免...
24育兒指導員到府可提供哪些服務內容？（一）提供嬰幼兒進食、遊戲、睡眠及幼兒活動路線等適性與安全諮詢。（二）提供嬰幼兒穿衣、換尿布...
25目前有領取育嬰留職停薪不能請領育兒津貼，但育嬰留職停薪只能請領六個月，之後孩子仍未滿2歲，還...領取育嬰留職停薪津貼的六個月內，同一孩童不得再領取育兒津貼。但如果在育嬰留職停薪期間但未領取...
26居家式托育人員(原保母)或親屬保母每人至多可以照顧幾位幼童？居家式托育人員或親屬保母每人至多照顧兒童(含托育人員本人之幼兒)4人，其中未滿2歲者最多2人...
27申請育兒指導服務要符合怎樣的資格？只要符合下列資格之一，即可申請：（一）設籍且實際居住本市，家有0-2歲以下幼兒並有育兒指導需...
28如何申請育兒指導服務？（一）、電話諮詢且評估到府指導需求，服務電話：03-5396000。（二）、新竹市政府社會處...
29育兒指導服務可以提供哪些協助？全部免費服務嗎？育兒指導服務是新竹市政府社會處主辦的免費服務方案；提供電話親子教養諮詢，且評估實際需求後，由...
.........
297取用食物的時候要注意什麼嗎？106-12-31
298取用食物有限制數量及區域嗎？106-12-31
299放在冰箱中的食物安全嗎?106-12-31
300續食是什麼?106-12-31
301什麼是愛享冰箱?106-12-31
302推動社區愛享冰箱的目的?106-12-31
303請問如何進行合作社之清算？合作社（場）於奉令解散或陳准解散登記後，應即進行清算手續。合作社（場）之清算登記，旨在使得合...
304如何申請本市社會福利慈善基金會？業務範圍僅在新竹市者，向新竹市政府社會處申請，其設立基金最低數額為一千萬元以上。應備文件（各...
305本市人民團體申請設立程序及法令依據？依據人民團體法第8條規定人民團體之組織，應由發起人檢具申請書、章程草案及發起人名冊，向主管機...
306請問有關合作社變更登記應備文件為何？1.變更登記申請書。2.入社社員名冊。3.退（出）社社員名冊。4.年度社員大會紀錄(含改選理...
307本市人民團體選舉相關重要法令？人民團體選舉罷免辦法第5條人民團體理事、監事及會員代表之選舉或罷免，應由理事會在召開會議十五...
308請問合作社場解散之程序為何？合作社因左列各款情事之一而解散：1.章程所定解散之事由發生。2.社員大會之解散決議（應有全體...
309請問如何申請籌組合作社場？有共同需要實行合作的人，如要組織合作社（場）時，應有七人以上發起人，發起前應先洽詢主管機關索...
310申請籌組人民團體之條件為何？一、申請條件：須設籍於本市之市民30人以上發起二、發起人之基本資格：須本市市民、年滿二十歲，...
311何謂合作社？所謂合作社係指依平等原則，在互助組織之基礎上，以共同經營方法，謀社員經濟之利益與生活之改善，...
312如何申請設立財團法人社會福利慈善事業基金會？一、應檢附下列文件：1.申請書(一份)2.基金會概況表(一份)3.基金會籌備會議記錄(四份)...
313申請人民團體應附書表及證件為何？一、社會團體：1、新竹市社會團體申請書2、新竹市社會團體章程3、新竹市社會團體發起人名冊4、...
314如何加入志工行列？參考本府所編製之志願服務資源簡章，瞭解各類別志工隊服務內容及所需志工資格，再依本身志趣，逕向...
315如何申請志工隊依志願服務法所訂志願服務運用單位係指運用志工之機關、機構、學校、法人或經政府立案之團體。符合...
316如何申請志願服務紀錄冊需先完成12小時之志願服務基礎訓練和特殊訓練。完成上述訓練者，由志願服務運用單位志願服務運用...
317如何申請志願服務榮譽卡志工服務年資滿3年、服務時數達300小時以上者，得檢具1吋半身照2張、服務紀錄冊影本、申請表...
318加入國民年金保險，經審查符合所得未達一定標準者，如有戶籍遷徙，是否應重新申請？國民年金保費減免係依被保險人當月底戶籍所在地直轄市、縣（市）主管機關為準。故申請符合所得未達...
319申請國民年金所得未達一定標準資格認定，應向哪一單位提出申請？請攜帶申請人身分證、印章及全戶戶口名簿等資料至戶籍所在地區公所社政課提出申請。
320國民年金開辦時年滿65歲之一般國民，請領老年基本保證年金之資格為何？國民年金開辦時年滿65歲國民，在國內設有戶籍，且於最近3年內每年居住超過183日，而無下列情...
321國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於...
322國民年金之實施對象為何？1.年滿25歲、未滿65歲，在國內設有戶籍，且未參加勞保、農保、公教保、軍保之國民。2.國民...
323國民年金生育給付從什麼時候開辦？自100年7月1日實施。
324國民年金生育給付從何時開始調漲？自104年12月18日起調漲，由1個月的國民年金月投保金額調整至2個月。105年月投保金額為...
325國民年金生育給付的申請資格？1.生產期間需有國民年金保險身分。2.自100年7月1日後生產。3.生產日起算5年內需向各地...
326國民年金所得未達一定標準保費減免的申請方式？1.向戶籍所在地區公所申請及洽詢。2.攜帶身分證、印章、戶口名簿。3.全家人口之退休俸、學生...
\n", + "

326 rows × 2 columns

\n", + "
" + ], + "text/plain": [ + " question \\\n", + "0 小孩出生後應於何時申請育兒津貼? \n", + "1 小孩出生後應於何時申請育兒津貼? \n", + "2 育兒津貼申請應備文件為何? \n", + "3 若民眾夫妻雙方均失業，是否可申請家庭育兒津貼費用補助 \n", + "4 育兒津貼補助對象為何？ \n", + "5 如何認定是否符合父母未就業家庭育兒津貼中的未就業？ \n", + "6 未就業家庭育兒津貼條件中，若一方無工作但有投保公會勞保，請問是否可申請該項津貼，請提供該項業... \n", + "7 就業者家庭部分托育費用補助所稱的「就業者」為何？ \n", + "8 我未來想當合格托育人員，但又不是相關科系，新竹市哪裡會開辦托育人員訓練課程呢？ \n", + "9 我該如何成為居家托育服務中心(原社區保母系統)的居家式托育人員呢？ \n", + "10 親戚朋友托育，是否也可以領補助呢？ \n", + "11 就業者家庭部分托育費用補助申請應備文件為何？向誰申請 \n", + "12 育兒津貼補助內容為為何？ \n", + "13 誰可以提出育兒津貼申請手續？ \n", + "14 托育費用補助標準為何？ \n", + "15 托育費用補助對象為何？ \n", + "16 育兒津貼補助金額多少？ \n", + "17 請提供本年度保母系統(居家托育服務中心)相關資訊為何? \n", + "18 育兒津貼受理單位為何？審核程序為何？ \n", + "19 要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？ \n", + "20 就業者家庭部分托育費用補助申請對象及審核程序為何？ \n", + "21 就業者家庭部分托育費用補助審核程序為何？ \n", + "22 托育補助金額多少，補助期間多長？ \n", + "23 免費育兒指導的內容是什麼？ \n", + "24 育兒指導員到府可提供哪些服務內容？ \n", + "25 目前有領取育嬰留職停薪不能請領育兒津貼，但育嬰留職停薪只能請領六個月，之後孩子仍未滿2歲，還... \n", + "26 居家式托育人員(原保母)或親屬保母每人至多可以照顧幾位幼童？ \n", + "27 申請育兒指導服務要符合怎樣的資格？ \n", + "28 如何申請育兒指導服務？ \n", + "29 育兒指導服務可以提供哪些協助？全部免費服務嗎？ \n", + ".. ... \n", + "297 取用食物的時候要注意什麼嗎？ \n", + "298 取用食物有限制數量及區域嗎？ \n", + "299 放在冰箱中的食物安全嗎? \n", + "300 續食是什麼? \n", + "301 什麼是愛享冰箱? \n", + "302 推動社區愛享冰箱的目的? \n", + "303 請問如何進行合作社之清算？ \n", + "304 如何申請本市社會福利慈善基金會？ \n", + "305 本市人民團體申請設立程序及法令依據？ \n", + "306 請問有關合作社變更登記應備文件為何？ \n", + "307 本市人民團體選舉相關重要法令？ \n", + "308 請問合作社場解散之程序為何？ \n", + "309 請問如何申請籌組合作社場？ \n", + "310 申請籌組人民團體之條件為何？ \n", + "311 何謂合作社？ \n", + "312 如何申請設立財團法人社會福利慈善事業基金會？ \n", + "313 申請人民團體應附書表及證件為何？ \n", + "314 如何加入志工行列？ \n", + "315 如何申請志工隊 \n", + "316 如何申請志願服務紀錄冊 \n", + "317 如何申請志願服務榮譽卡 \n", + "318 加入國民年金保險，經審查符合所得未達一定標準者，如有戶籍遷徙，是否應重新申請？ \n", + "319 申請國民年金所得未達一定標準資格認定，應向哪一單位提出申請？ \n", + "320 國民年金開辦時年滿65歲之一般國民，請領老年基本保證年金之資格為何？ \n", + "321 國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？ \n", + "322 國民年金之實施對象為何？ \n", + "323 國民年金生育給付從什麼時候開辦？ \n", + "324 國民年金生育給付從何時開始調漲？ \n", + "325 國民年金生育給付的申請資格？ \n", + "326 國民年金所得未達一定標準保費減免的申請方式？ \n", + "\n", + " ans \n", + "0 1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在... \n", + "1 隨時提出;津貼經審查通過後，追溯自受理申請之當月起發給。兒童出生後六十日內向戶政事務所完成出... \n", + "2 申請資料應備齊:(一)兒童之戶口名簿影本。(二)申請人之郵局存摺封面影本。(三)父母雙方身分... \n", + "3 一、育兒津貼補助對象：1.育有二足歲以下兒童。2.兒童之父母至少一方因育兒需要，致未能就業者... \n", + "4 育兒津貼補助對象，應符合下列規定：(一)育有二足歲以下兒童。(二)兒童之父母(或監護人)至少... \n", + "5 1.未參加勞工「就業」保險（或軍保、公保）。2.最近一年度之稅捐稽徵機關核定之薪資所得及執行... \n", + "6 針對勞工保險投保職業工會者（或雇主）等對象，一般均未參加「勞工就業保險」（自行加保者除外），... \n", + "7 (1)受僱於政府、學校或公民營事業單位者。(2)勞動基準法所稱受雇主僱用從事工作者。(3)依... \n", + "8 勞動部勞動力發展署及本府委託社團法人新竹市社區婦女關懷協會、社團法人新竹市嬰幼兒保育協會、新... \n", + "9 欲申請加入居家托育服務中心內的居家托育人員依規需年滿20歲並符合以下資格之一：1.取得保母技... \n", + "10 一般民眾(含寶寶的爺爺奶奶、阿姨、叔叔、舅舅等三親等以內親屬)必須上課126小時托育人員訓練... \n", + "11 1.補助申請表。2.最近戶籍謄本或戶口名簿影本。3.郵局存簿封面影本。4.托育契約書（親屬托... \n", + "12 本津貼補助對象，應符合下列規定：(一)育有2足歲以下兒童。(二)兒童之父母(或監護人)至少一... \n", + "13 (一)申請人資格規定如下：1.育有二足歲以下兒童。2.兒童之父母(或監護人)至少一方因育兒需... \n", + "14 補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托... \n", + "15 補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總... \n", + "16 育兒津貼補助金額依家庭狀況而定，補助金額如下：(一)低收入戶：每名兒童每月補助新臺幣5000... \n", + "17 目前本市居家托育服務中心承辦單位及聯絡電話如下：東區居家托育服務中心承辦單位：新竹市私立光復... \n", + "18 本津貼申領及發放程序規定如下：(一)由申請人檢具相關證明文件向兒童戶籍地之區公所提出申請。(... \n", + "19 依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服... \n", + "20 就業者家庭部分托育費用補助的實施對象為：1.父母(或監護人)雙方或單親一方皆就業，或父母一方... \n", + "21 父母需於托育事實發生日起15日內，檢齊應備文件送交幼兒托育地點之社區保母系統或托嬰中心，其補... \n", + "22 依各類家庭條件及送托托育人員資格，其托育補助金額說明如下：(一)將嬰幼兒送至托育服務中心中有... \n", + "23 政府建構友善的托育環境，除了提供托育費用及育兒津貼的經濟補助外，仍然重視親職教育，所以提供免... \n", + "24 （一）提供嬰幼兒進食、遊戲、睡眠及幼兒活動路線等適性與安全諮詢。（二）提供嬰幼兒穿衣、換尿布... \n", + "25 領取育嬰留職停薪津貼的六個月內，同一孩童不得再領取育兒津貼。但如果在育嬰留職停薪期間但未領取... \n", + "26 居家式托育人員或親屬保母每人至多照顧兒童(含托育人員本人之幼兒)4人，其中未滿2歲者最多2人... \n", + "27 只要符合下列資格之一，即可申請：（一）設籍且實際居住本市，家有0-2歲以下幼兒並有育兒指導需... \n", + "28 （一）、電話諮詢且評估到府指導需求，服務電話：03-5396000。（二）、新竹市政府社會處... \n", + "29 育兒指導服務是新竹市政府社會處主辦的免費服務方案；提供電話親子教養諮詢，且評估實際需求後，由... \n", + ".. ... \n", + "297 106-12-31 \n", + "298 106-12-31 \n", + "299 106-12-31 \n", + "300 106-12-31 \n", + "301 106-12-31 \n", + "302 106-12-31 \n", + "303 合作社（場）於奉令解散或陳准解散登記後，應即進行清算手續。合作社（場）之清算登記，旨在使得合... \n", + "304 業務範圍僅在新竹市者，向新竹市政府社會處申請，其設立基金最低數額為一千萬元以上。應備文件（各... \n", + "305 依據人民團體法第8條規定人民團體之組織，應由發起人檢具申請書、章程草案及發起人名冊，向主管機... \n", + "306 1.變更登記申請書。2.入社社員名冊。3.退（出）社社員名冊。4.年度社員大會紀錄(含改選理... \n", + "307 人民團體選舉罷免辦法第5條人民團體理事、監事及會員代表之選舉或罷免，應由理事會在召開會議十五... \n", + "308 合作社因左列各款情事之一而解散：1.章程所定解散之事由發生。2.社員大會之解散決議（應有全體... \n", + "309 有共同需要實行合作的人，如要組織合作社（場）時，應有七人以上發起人，發起前應先洽詢主管機關索... \n", + "310 一、申請條件：須設籍於本市之市民30人以上發起二、發起人之基本資格：須本市市民、年滿二十歲，... \n", + "311 所謂合作社係指依平等原則，在互助組織之基礎上，以共同經營方法，謀社員經濟之利益與生活之改善，... \n", + "312 一、應檢附下列文件：1.申請書(一份)2.基金會概況表(一份)3.基金會籌備會議記錄(四份)... \n", + "313 一、社會團體：1、新竹市社會團體申請書2、新竹市社會團體章程3、新竹市社會團體發起人名冊4、... \n", + "314 參考本府所編製之志願服務資源簡章，瞭解各類別志工隊服務內容及所需志工資格，再依本身志趣，逕向... \n", + "315 依志願服務法所訂志願服務運用單位係指運用志工之機關、機構、學校、法人或經政府立案之團體。符合... \n", + "316 需先完成12小時之志願服務基礎訓練和特殊訓練。完成上述訓練者，由志願服務運用單位志願服務運用... \n", + "317 志工服務年資滿3年、服務時數達300小時以上者，得檢具1吋半身照2張、服務紀錄冊影本、申請表... \n", + "318 國民年金保費減免係依被保險人當月底戶籍所在地直轄市、縣（市）主管機關為準。故申請符合所得未達... \n", + "319 請攜帶申請人身分證、印章及全戶戶口名簿等資料至戶籍所在地區公所社政課提出申請。 \n", + "320 國民年金開辦時年滿65歲國民，在國內設有戶籍，且於最近3年內每年居住超過183日，而無下列情... \n", + "321 國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於... \n", + "322 1.年滿25歲、未滿65歲，在國內設有戶籍，且未參加勞保、農保、公教保、軍保之國民。2.國民... \n", + "323 自100年7月1日實施。 \n", + "324 自104年12月18日起調漲，由1個月的國民年金月投保金額調整至2個月。105年月投保金額為... \n", + "325 1.生產期間需有國民年金保險身分。2.自100年7月1日後生產。3.生產日起算5年內需向各地... \n", + "326 1.向戶籍所在地區公所申請及洽詢。2.攜帶身分證、印章、戶口名簿。3.全家人口之退休俸、學生... \n", + "\n", + "[326 rows x 2 columns]" + ] + }, + "execution_count": 55, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# 把檔案讀出來(原始資料: https://society.hccg.gov.tw/ch/home.jsp?id=43&parentpath=0,5)\n", + "df_QA = pd.read_json('ProcessedData.json', encoding='utf8')\n", + "# 我們這次只會使用到question跟ans這兩個欄位\n", + "df_question = df_QA[['question', 'ans']].copy() ## 不要更動到原始的DataFrame\n", + "df_question.drop_duplicates(inplace=True) ## 丟掉重複的資料\n", + "df_question ## show出來" + ] + }, + { + "cell_type": "code", + "execution_count": 56, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Building prefix dict from D:\\Projects\\IIIMaterial\\08_IrTheory\\dict.txt.big ...\n", + "Loading model from cache C:\\Users\\user\\AppData\\Local\\Temp\\jieba.u113aa1b94b8c59a47d44fb9a8d37f14b.cache\n", + "Loading model cost 1.658 seconds.\n", + "Prefix dict has been built succesfully.\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "question 小孩出生後應於何時申請育兒津貼?\n", + "ans 1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在...\n", + "processed [小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ]\n", + "Name: 0, dtype: object\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
questionansprocessed
0小孩出生後應於何時申請育兒津貼?1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在...[小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ]
1小孩出生後應於何時申請育兒津貼?隨時提出;津貼經審查通過後，追溯自受理申請之當月起發給。兒童出生後六十日內向戶政事務所完成出...[小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ]
2育兒津貼申請應備文件為何?申請資料應備齊:(一)兒童之戶口名簿影本。(二)申請人之郵局存摺封面影本。(三)父母雙方身分...[育兒, 津貼, 申請, 應, 備, 文件, 為, 何, , ]
3若民眾夫妻雙方均失業，是否可申請家庭育兒津貼費用補助一、育兒津貼補助對象：1.育有二足歲以下兒童。2.兒童之父母至少一方因育兒需要，致未能就業者...[若, 民, 眾, 夫妻, 雙方, 均, 失業, , , 是否, 可, 申請, 家庭, 育兒...
4育兒津貼補助對象為何？育兒津貼補助對象，應符合下列規定：(一)育有二足歲以下兒童。(二)兒童之父母(或監護人)至少...[育兒, 津貼, 貼補, 補助, 對象, 為, 何, , ]
\n", + "
" + ], + "text/plain": [ + " question \\\n", + "0 小孩出生後應於何時申請育兒津貼? \n", + "1 小孩出生後應於何時申請育兒津貼? \n", + "2 育兒津貼申請應備文件為何? \n", + "3 若民眾夫妻雙方均失業，是否可申請家庭育兒津貼費用補助 \n", + "4 育兒津貼補助對象為何？ \n", + "\n", + " ans \\\n", + "0 1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在... \n", + "1 隨時提出;津貼經審查通過後，追溯自受理申請之當月起發給。兒童出生後六十日內向戶政事務所完成出... \n", + "2 申請資料應備齊:(一)兒童之戶口名簿影本。(二)申請人之郵局存摺封面影本。(三)父母雙方身分... \n", + "3 一、育兒津貼補助對象：1.育有二足歲以下兒童。2.兒童之父母至少一方因育兒需要，致未能就業者... \n", + "4 育兒津貼補助對象，應符合下列規定：(一)育有二足歲以下兒童。(二)兒童之父母(或監護人)至少... \n", + "\n", + " processed \n", + "0 [小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ] \n", + "1 [小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ] \n", + "2 [育兒, 津貼, 申請, 應, 備, 文件, 為, 何, , ] \n", + "3 [若, 民, 眾, 夫妻, 雙方, 均, 失業, , , 是否, 可, 申請, 家庭, 育兒... \n", + "4 [育兒, 津貼, 貼補, 補助, 對象, 為, 何, , ] " + ] + }, + "execution_count": 56, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#前處理\n", + "all_terms = []\n", + "def preprocess(item): ##定義前處理的function\n", + " # 請把將每一行用jieba.cut進行分詞(記得將cut_all設定為True)\n", + " # 同時建立所有詞彙的list(all_terms)\n", + " #=============your works starts===============#\n", + " terms = [t for t in jieba.cut(item, cut_all=True)] ## 把全切分模式打開，可以比對的詞彙比較多\n", + " all_terms.extend(terms) ## 收集所有出現過的字\n", + " #==============your works ends================#\n", + " return terms\n", + "\n", + "df_question['processed'] = df_question['question'].apply(preprocess)\n", + "print(df_question.iloc[0])\n", + "# question 小孩出生後應於何時申請育兒津貼?\n", + "# ans 1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在...\n", + "# processed [小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ]\n", + "# Name: 0, dtype: object\n", + "\n", + "df_question.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 65, + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "len(termindex) 1012\n", + "['', '耗材', '被', '其他', '發', '發現', '申請人', '遭遇', '環境', '您好']\n" + ] + } + ], + "source": [ + "# 建立termindex: 將all_terms取出不重複的詞彙，並轉換型別為list(避免順序亂掉)\n", + "#=============your works starts===============#\n", + "termindex = list(set(all_terms))\n", + "#==============your works ends================#\n", + "\n", + "print(\"len(termindex)\", len(termindex))\n", + "print(termindex[:10])\n", + "# len(termindex) 1012\n", + "# ['', '耗材', '被', '其他', '發', '發現', '申請人', '遭遇', '環境', '您好']" + ] + }, + { + "cell_type": "code", + "execution_count": 63, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "1012\n", + "[0.04710446918747347, 5.786897381366708, 5.786897381366708, 5.786897381366708, 4.400603020246817, 4.688285092698598, 5.093750200806762, 5.093750200806762, 5.786897381366708, 5.786897381366708]\n" + ] + } + ], + "source": [ + "# 建立IDF vector\n", + "Doc_Length = len(df_question) ## 計算出共有幾篇文章\n", + "Idf_vector = [] ## 初始化IDF向量\n", + "for term in termindex: ## 對index中的詞彙跑回圈\n", + " num_of_doc_contains_term = 0 ## 計算有機篇文章出現過這個詞彙\n", + " for terms in df_question['processed']:\n", + " if term in terms:\n", + " num_of_doc_contains_term += 1\n", + " idf = np.log(Doc_Length/num_of_doc_contains_term) ## 計算該詞彙的IDF值\n", + " Idf_vector.append(idf)\n", + "print(len(Idf_vector))\n", + "print(Idf_vector[:10])" + ] + }, + { + "cell_type": "code", + "execution_count": 68, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "0 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "1 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "2 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "3 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "4 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "5 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "6 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "7 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "8 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "9 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "Name: vector, dtype: object" + ] + }, + "execution_count": 68, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# 建立document vector\n", + "def terms_to_vector(terms): ## 定義把terms轉換成向量的function\n", + " ## 建立一條與termsindex等長、但值全部為零的向量(hint:dtype=np.float32)\n", + " #=============your works starts===============#\n", + " vector = np.zeros_like(termindex, dtype=np.float32) \n", + " #==============your works ends================#\n", + " \n", + " for term, count in Counter(terms).items():\n", + " # 計算vector上每一個字的tf值\n", + " #=============your works starts===============#\n", + " vector[termindex.index(term)] = count\n", + " #==============your works ends================#\n", + "\n", + " # 計算tfidf，element-wise的將vector與Idf_vector相乘\n", + " ## hint: 如果兩個vector的型別都是np.array，把兩條vector相乘，就會自動把向量中的每一個元素成在一起，建立出一條新的向量\n", + " #=============your works starts===============#\n", + " vector = vector * Idf_vector\n", + " #==============your works ends================#\n", + " return vector\n", + "\n", + "\n", + "\n", + "df_question['vector'] = df_question['processed'].apply(terms_to_vector) ## 將上面定義的function，套用在每一筆資料的terms欄位上\n", + "df_question['vector'][:10]\n", + "# 0 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "# 1 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "# 2 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "# 3 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "# 4 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "# 5 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "# 6 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "# 7 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "# 8 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "# 9 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...\n", + "# Name: vector, dtype: object" + ] + }, + { + "cell_type": "code", + "execution_count": 69, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "小孩出生後應於何時申請育兒津貼?\n", + "育兒津貼申請應備文件為何?\n", + "0.203227847937731\n" + ] + } + ], + "source": [ + "from numpy.linalg import norm\n", + "\n", + "def cosine_similarity(vector1, vector2): ## 定義cosine相似度的計算公式\n", + " # 使用np.dot與norm計算cosine score\n", + " #=============your works starts===============#\n", + " score = np.dot(vector1, vector2) / (norm(vector1) * norm(vector2))\n", + " #==============your works ends================#\n", + " return score\n", + "\n", + "sentence1 = df_question.loc[0] ##取出第零個的問題\n", + "sentence2 = df_question.loc[2] ##取出第二個的問題\n", + "print(sentence1['question'])\n", + "print(sentence2['question'])\n", + "print(cosine_similarity(sentence1['vector'], sentence2['vector'])) ##計算兩者的相似度\n", + "# 0.203227847937731" + ] + }, + { + "cell_type": "code", + "execution_count": 88, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Float64Index([100.0, 111.0, 321.0], dtype='float64')\n" + ] + } + ], + "source": [ + "def retrieve(testing_sentence, return_num=3): ## 定義出檢索引擎\n", + " # 請使用前面定義的terms_to_vector與preprocess兩個function，計算出testing_sentence的向量\n", + " # 計算其與資料庫每一的問句的相似度\n", + " # 依分數進行排序，找到分數最高的三個句子\n", + " #=============your works starts===============#\n", + " testing_vector = terms_to_vector(preprocess(testing_sentence)) ## 把剛剛的前處理、轉換成向量的function，應用在使用者輸入的問題上\n", + " idx_score_mapping = [(idx, cosine_similarity(testing_vector, vec)) for idx, vec in enumerate(df_question['vector'])]\n", + " top3_idxs = np.array(sorted(idx_score_mapping, key=lambda x:x[1], reverse=True))[:3, 0]\n", + " #==============your works ends================#\n", + " \n", + " return df_question.loc[top3_idxs, ['question', 'ans']]\n", + "\n", + "\n", + "print(retrieve(\"老人年金\").index)\n", + "# Float64Index([100.0, 111.0, 321.0], dtype='float64')\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Use Scikit learn" + ] + }, + { + "cell_type": "code", + "execution_count": 94, + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn.feature_extraction.text import TfidfVectorizer" + ] + }, + { + "cell_type": "code", + "execution_count": 136, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[2.54619627 2.54619627 1.95695906 3.12409736 2.19106254 2.74144953\n", + " 3.82923767 2.54569516 3.4163518 2.98088982 2.35528293]\n" + ] + } + ], + "source": [ + "tfidf = TfidfVectorizer()\n", + "# 使用tfidf.fit_transform將轉換df_question['processed']為vector\n", + "#=============your works starts===============#\n", + "df_question['sklearn_vector'] = list(tfidf.fit_transform(df_question['processed'].apply(lambda x:\" \".join(x)).values).toarray())\n", + "#==============your works ends================#\n", + "\n", + "print(df_question.loc[:10, 'sklearn_vector'].apply(sum).values)\n", + "# [2.54619627 2.54619627 1.95695906 3.12409736 2.19106254 2.74144953\n", + "# 3.82923767 2.54569516 3.4163518 2.98088982 2.35528293]" + ] + }, + { + "cell_type": "code", + "execution_count": 137, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "100.0 我已經年滿65歲領有國民年金老人年金及基本保證年金3628元，因家境清寒還可以再申請中低收入...\n", + "111.0 新竹市老人一般可領老人津貼6628元，該如何申請？\n", + "321.0 國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？\n", + "Name: question, dtype: object\n", + "100.0 我已經年滿65歲領有國民年金老人年金及基本保證年金3628元，因家境清寒還可以再申請中低收入...\n", + "111.0 新竹市老人一般可領老人津貼6628元，該如何申請？\n", + "321.0 國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？\n", + "Name: question, dtype: object\n" + ] + } + ], + "source": [ + "def sklearn_retrieve(testing_sentence, return_num=3): ## 定義出檢索引擎\n", + " # 請使用前面定義的tfidf.transform與preprocess兩個function，計算出testing_sentence的向量\n", + " # 注意tfidf.transform必須是兩個維度的array\n", + " # 且out為sparse metric，必需.toarray()轉換為一般np.array()\n", + " # 計算其與資料庫每一的問句的相似度\n", + " # 依分數進行排序，找到分數最高的三個句子\n", + " #=============your works starts===============#\n", + " testing_vector = tfidf.transform([\" \".join(preprocess(testing_sentence))]).toarray()[0]\n", + " idx_score_mapping = [(idx, cosine_similarity(testing_vector, vec)) for idx, vec in enumerate(df_question['sklearn_vector'])]\n", + " top3_idxs = np.array(sorted(idx_score_mapping, key=lambda x:x[1], reverse=True))[:3, 0]\n", + " #==============your works ends================#\n", + " return df_question.loc[top3_idxs, ['question', 'ans']]\n", + "\n", + "print(retrieve(\"老人年金\")['question'])\n", + "print(sklearn_retrieve(\"老人年金\")['question'])\n", + "# 100.0 我已經年滿65歲領有國民年金老人年金及基本保證年金3628元，因家境清寒還可以再申請中低收入...\n", + "# 111.0 新竹市老人一般可領老人津貼6628元，該如何申請？\n", + "# 321.0 國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？\n", + "# Name: question, dtype: object\n", + "# 100.0 我已經年滿65歲領有國民年金老人年金及基本保證年金3628元，因家境清寒還可以再申請中低收入...\n", + "# 111.0 新竹市老人一般可領老人津貼6628元，該如何申請？\n", + "# 321.0 國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？\n", + "# Name: question, dtype: object" + ] + }, + { + "cell_type": "code", + "execution_count": 138, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "15.0 托育費用補助對象為何？\n", + "14.0 托育費用補助標準為何？\n", + "19.0 要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？\n", + "Name: question, dtype: object\n", + "0.0 小孩出生後應於何時申請育兒津貼?\n", + "1.0 小孩出生後應於何時申請育兒津貼?\n", + "2.0 育兒津貼申請應備文件為何?\n", + "Name: question, dtype: object\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "c:\\users\\user\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\ipykernel_launcher.py:6: RuntimeWarning: invalid value encountered in double_scalars\n", + " \n" + ] + } + ], + "source": [ + "print(retrieve(\"托育\")['question'])\n", + "print(sklearn_retrieve(\"托育\")['question'])\n" + ] + }, + { + "cell_type": "code", + "execution_count": 139, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "214.0 中低收入醫療補助補助項目及標準為何？\n", + "108.0 申請假牙補助的資格及補助內容\n", + "82.0 特殊境遇家庭法律訴訟補助如何申請？補助額度如何？\n", + "Name: question, dtype: object\n", + "108.0 申請假牙補助的資格及補助內容\n", + "15.0 托育費用補助對象為何？\n", + "214.0 中低收入醫療補助補助項目及標準為何？\n", + "Name: question, dtype: object\n" + ] + } + ], + "source": [ + "print(retrieve(\"補助\")['question'])\n", + "print(sklearn_retrieve(\"補助\")['question'])\n" + ] + }, + { + "cell_type": "code", + "execution_count": 140, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "204.0 社會福利-急難救助核發救助對象？\n", + "74.0 遭遇特殊境遇家庭如何申請救助(申請方式)？\n", + "203.0 社會福利-我要到那裡申請急難救助？\n", + "Name: question, dtype: object\n", + "204.0 社會福利-急難救助核發救助對象？\n", + "203.0 社會福利-我要到那裡申請急難救助？\n", + "74.0 遭遇特殊境遇家庭如何申請救助(申請方式)？\n", + "Name: question, dtype: object\n" + ] + } + ], + "source": [ + "print(retrieve(\"救助\")['question'])\n", + "print(sklearn_retrieve(\"救助\")['question'])\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.6.4" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/08_IrTheory/simsun.ttf b/08_IrTheory/simsun.ttf new file mode 100644 index 0000000..e0115ab Binary files /dev/null and b/08_IrTheory/simsun.ttf differ diff --git a/08_IrTheory/stops.txt b/08_IrTheory/stops.txt new file mode 100644 index 0000000..ac5b4df --- /dev/null +++ b/08_IrTheory/stops.txt @@ -0,0 +1,1212 @@ +, + +、 +。 +“ +” +《 +》 +！ +， +： +； +？ +人民 +末##末 +啊 +阿 +哎 +哎呀 +哎喲 +唉 +我 +我們 +按 +按照 +依照 +吧 +吧噠 +把 +罷了 +被 +本 +本著 +比 +比方 +比如 +鄙人 +彼 +彼此 +邊 +別 +別的 +別說 +並 +並且 +不比 +不成 +不單 +不但 +不獨 +不管 +不光 +不過 +不僅 +不拘 +不論 +不怕 +不然 +不如 +不特 +不惟 +不問 +不只 +朝 +朝著 +趁 +趁著 +乘 +沖 +除 +除此之外 +除非 +除了 +此 +此間 +此外 +從 +從而 +打 +待 +但 +但是 +當 +當著 +到 +得 +的 +的話 +等 +等等 +地 +第 +叮咚 +對 +對於 +多 +多少 +而 +而況 +而且 +而是 +而外 +而言 +而已 +爾後 +反過來 +反過來說 +反之 +非但 +非徒 +否則 +嘎 +嘎登 +該 +趕 +個 +各 +各個 +各位 +各種 +各自 +給 +根據 +跟 +故 +故此 +固然 +關於 +管 +歸 +果然 +果真 +過 +哈 +哈哈 +呵 +和 +何 +何處 +何況 +何時 +嘿 +哼 +哼唷 +呼哧 +乎 +嘩 +還是 +還有 +換句話說 +換言之 +或 +或是 +或者 +極了 +及 +及其 +及至 +即 +即便 +即或 +即令 +即若 +即使 +幾 +幾時 +己 +既 +既然 +既是 +繼而 +加之 +假如 +假若 +假使 +鑒於 +將 +較 +較之 +叫 +接著 +結果 +借 +緊接著 +進而 +盡 +儘管 +經 +經過 +就 +就是 +就是說 +據 +具體地說 +具體說來 +開始 +開外 +靠 +咳 +可 +可見 +可是 +可以 +況且 +啦 +來 +來著 +離 +例如 +哩 +連 +連同 +兩者 +了 +臨 +另 +另外 +另一方面 +論 +嘛 +嗎 +慢說 +漫說 +冒 +麼 +每 +每當 +們 +莫若 +某 +某個 +某些 +拿 +哪 +哪邊 +哪兒 +哪個 +哪裏 +哪年 +哪怕 +哪天 +哪些 +哪樣 +那 +那邊 +那兒 +那個 +那會兒 +那裏 +那麼 +那麼些 +那麼樣 +那時 +那些 +那樣 +乃 +乃至 +呢 +能 +你 +你們 +您 +寧 +寧可 +寧肯 +寧願 +哦 +嘔 +啪達 +旁人 +呸 +憑 +憑藉 +其 +其次 +其二 +其他 +其它 +其一 +其餘 +其中 +起 +起見 +豈但 +恰恰相反 +前後 +前者 +且 +然而 +然後 +然則 +讓 +人家 +任 +任何 +任憑 +如 +如此 +如果 +如何 +如其 +如若 +如上所述 +若 +若非 +若是 +啥 +上下 +尚且 +設若 +設使 +甚而 +甚麼 +甚至 +省得 +時候 +什麼 +什麼樣 +使得 +是 +是的 +首先 +誰 +誰知 +順 +順著 +似的 +雖 +雖然 +雖說 +雖則 +隨 +隨著 +所 +所以 +他 +他們 +他人 +它 +它們 +她 +她們 +倘 +倘或 +倘然 +倘若 +倘使 +騰 +替 +通過 +同 +同時 +哇 +萬一 +往 +望 +為 +為何 +為了 +為什麼 +為著 +餵 +嗡嗡 +我 +我們 +嗚 +嗚呼 +烏乎 +無論 +無寧 +毋寧 +嘻 +嚇 +相對而言 +像 +向 +向著 +噓 +呀 +焉 +沿 +沿著 +要 +要不 +要不然 +要不是 +要麼 +要是 +也 +也罷 +也好 +一 +一般 +一旦 +一方面 +一來 +一切 +一樣 +一則 +依 +依照 +矣 +以 +以便 +以及 +以免 +以至 +以至於 +以致 +抑或 +因 +因此 +因而 +因為 +喲 +用 +由 +由此可見 +由於 +有 +有的 +有關 +有些 +又 +於 +於是 +於是乎 +與 +與此同時 +與否 +與其 +越是 +雲雲 +哉 +再說 +再者 +在 +在下 +咱 +咱們 +則 +怎 +怎麼 +怎麼辦 +怎麼樣 +怎樣 +咋 +照 +照著 +者 +這 +這邊 +這兒 +這個 +這會兒 +這就是說 +這裏 +這麼 +這麼點兒 +這麼些 +這麼樣 +這時 +這些 +這樣 +正如 +吱 +之 +之類 +之所以 +之一 +只是 +只限 +只要 +只有 +至 +至於 +諸位 +著 +著呢 +自 +自從 +自個兒 +自各兒 +自己 +自家 +自身 +綜上所述 +總的來看 +總的來說 +總的說來 +總而言之 +總之 +縱 +縱令 +縱然 +縱使 +遵照 +作為 +兮 +呃 +唄 +咚 +咦 +喏 +啐 +喔唷 +嗬 +嗯 +噯 +~ +! +. + + +' +( +) + +A +白 +社會主義 +-- +.. + + [ + ] + + + + + + +- +_ ++ += +& +^ +% +# +@ +` +; +$ +（ +） +—— +— +￥ +· +... +‘ +’ +〉 +〈 +… +　 +0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +０ +１ +２ +３ +４ +５ +６ +７ +８ +９ +二 +三 +四 +五 +六 +七 +八 +九 +零 +＞ +＜ +＠ +＃ +＄ +％ +︿ +＆ +＊ +＋ +～ +｜ +［ +］ +｛ +｝ +啊哈 +啊呀 +啊喲 +挨次 +挨個 +挨家挨戶 +挨門挨戶 +挨門逐戶 +挨著 +按理 +按期 +按時 +按說 +暗地裏 +暗中 +暗自 +昂然 +八成 +白白 +半 +梆 +保管 +保險 +飽 +背地裏 +背靠背 +倍感 +倍加 +本人 +本身 +甭 +比起 +比如說 +比照 +畢竟 +必 +必定 +必將 +必須 +便 +別人 +並非 +並肩 +並沒 +並沒有 +併排 +並無 +勃然 +不 +不必 +不常 +不大 +不但...而且 +不得 +不得不 +不得了 +不得已 +不迭 +不定 +不對 +不妨 +不管怎樣 +不會 +不僅...而且 +不僅僅 +不僅僅是 +不經意 +不可開交 +不可抗拒 +不力 +不了 +不料 +不滿 +不免 +不能不 +不起 +不巧 +不然的話 +不日 +不少 +不勝 +不時 +不是 +不同 +不能 +不要 +不外 +不外乎 +不下 +不限 +不消 +不已 +不亦樂乎 +不由得 +不再 +不擇手段 +不怎麼 +不曾 +不知不覺 +不止 +不止一次 +不至於 +才 +才能 +策略地 +差不多 +差一點 +常 +常常 +常言道 +常言說 +常言說得好 +長此下去 +長話短說 +長期以來 +長線 +敞開兒 +徹夜 +陳年 +趁便 +趁機 +趁熱 +趁勢 +趁早 +成年 +成年累月 +成心 +乘機 +乘勝 +乘勢 +乘隙 +乘虛 +誠然 +遲早 +充分 +充其極 +充其量 +抽冷子 +臭 +初 +出 +出來 +出去 +除此 +除此而外 +除此以外 +除開 +除去 +除卻 +除外 +處處 +川流不息 +傳 +傳說 +傳聞 +串列 +純 +純粹 +此後 +此中 +次第 +匆匆 +從不 +從此 +從此以後 +從古到今 +從古至今 +從今以後 +從寬 +從來 +從輕 +從速 +從頭 +從未 +從無到有 +從小 +從新 +從嚴 +從優 +從早到晚 +從中 +從重 +湊巧 +粗 +存心 +達旦 +打從 +打開天窗說亮話 +大 +大不了 +大大 +大抵 +大都 +大多 +大凡 +大概 +大家 +大舉 +大略 +大面兒上 +大事 +大體 +大體上 +大約 +大張旗鼓 +大致 +呆呆地 +帶 +殆 +待到 +單 +單純 +單單 +但願 +彈指之間 +當場 +當兒 +當即 +當口兒 +當然 +當庭 +當頭 +當下 +當真 +當中 +倒不如 +倒不如說 +倒是 +到處 +到底 +到了兒 +到目前為止 +到頭 +到頭來 +得起 +得天獨厚 +的確 +等到 +叮噹 +頂多 +定 +動不動 +動輒 +陡然 +都 +獨 +獨自 +斷然 +頓時 +多次 +多多 +多多少少 +多多益善 +多虧 +多年來 +多年前 +而後 +而論 +而又 +爾等 +二話不說 +二話沒說 +反倒 +反倒是 +反而 +反手 +反之亦然 +反之則 +方 +方才 +方能 +放量 +非常 +非得 +分期 +分期分批 +分頭 +奮勇 +憤然 +風雨無阻 +逢 +弗 +甫 +嘎嘎 +該當 +概 +趕快 +趕早不趕晚 +敢 +敢情 +敢於 +剛 +剛才 +剛好 +剛巧 +高低 +格外 +隔日 +隔夜 +個人 +各式 +更 +更加 +更進一步 +更為 +公然 +共 +共總 +夠瞧的 +姑且 +古來 +故而 +故意 +固 +怪 +怪不得 +慣常 +光 +光是 +歸根到底 +歸根結底 +過於 +毫不 +毫無 +毫無保留地 +毫無例外 +好在 +何必 +何嘗 +何妨 +何苦 +何樂而不為 +何須 +何止 +很 +很多 +很少 +轟然 +後來 +呼啦 +忽地 +忽然 +互 +互相 +嘩啦 +話說 +還 +恍然 +會 +豁然 +活 +夥同 +或多或少 +或許 +基本 +基本上 +基於 +極 +極大 +極度 +極端 +極力 +極其 +極為 +急匆匆 +即將 +即刻 +即是說 +幾度 +幾番 +幾乎 +幾經 +既...又 +繼之 +加上 +加以 +間或 +簡而言之 +簡言之 +簡直 +見 +將才 +將近 +將要 +交口 +較比 +較為 +接連不斷 +接下來 +皆可 +截然 +截至 +藉以 +藉此 +藉以 +屆時 +僅 +僅僅 +謹 +進來 +進去 +近 +近幾年來 +近來 +近年來 +儘管如此 +儘可能 +儘快 +儘量 +盡然 +盡如人意 +盡心竭力 +盡心盡力 +儘早 +精光 +經常 +竟 +竟然 +究竟 +就此 +就地 +就算 +居然 +局外 +舉凡 +據稱 +據此 +據實 +據說 +據我所知 +據悉 +具體來說 +決不 +決非 +絕 +絕不 +絕頂 +絕對 +絕非 +均 +喀 +看 +看來 +看起來 +看上去 +看樣子 +可好 +可能 +恐怕 +快 +快要 +來不及 +來得及 +來講 +來看 +攔腰 +牢牢 +老 +老大 +老老實實 +老是 +累次 +累年 +理當 +理該 +理應 +歷 +立 +立地 +立刻 +立馬 +立時 +聯袂 +連連 +連日 +連日來 +連聲 +連袂 +臨到 +另方面 +另行 +另一個 +路經 +屢 +屢次 +屢次三番 +屢屢 +縷縷 +率爾 +率然 +略 +略加 +略微 +略為 +論說 +馬上 +蠻 +滿 +沒 +沒有 +每逢 +每每 +每時每刻 +猛然 +猛然間 +莫 +莫不 +莫非 +莫如 +默默地 +默然 +吶 +那末 +奈 +難道 +難得 +難怪 +難說 +內 +年覆一年 +凝神 +偶而 +偶爾 +怕 +砰 +碰巧 +譬如 +偏偏 +乒 +平素 +頗 +迫於 +撲通 +其後 +其實 +奇 +齊 +起初 +起來 +起首 +起頭 +起先 +豈 +豈非 +豈止 +迄 +恰逢 +恰好 +恰恰 +恰巧 +恰如 +恰似 +千 +千萬 +千萬千萬 +切 +切不可 +切莫 +切切 +切勿 +竊 +親口 +親身 +親手 +親眼 +親自 +頃 +頃刻 +頃刻間 +頃刻之間 +請勿 +窮年累月 +取道 +去 +權時 +全都 +全力 +全年 +全然 +全身心 +然 +人人 +仍 +仍舊 +仍然 +日覆一日 +日見 +日漸 +日益 +日臻 +如常 +如此等等 +如次 +如今 +如期 +如前所述 +如上 +如下 +汝 +三番兩次 +三番五次 +三天兩頭 +瑟瑟 +沙沙 +上 +上來 +上去 +\n +「 +」 \ No newline at end of file diff --git a/09_IrImplementation/day23_IRImplementation.ipynb b/09_IrImplementation/day23_IRImplementation.ipynb deleted file mode 100644 index 7809d5e..0000000 --- a/09_IrImplementation/day23_IRImplementation.ipynb +++ /dev/null @@ -1,2448 +0,0 @@ -{ - "cells": [ - { - "cell_type": "code", - "execution_count": 1, - "metadata": {}, - "outputs": [], - "source": [ - "import jieba\n", - "jieba.set_dictionary('dict.txt.big')\n", - "import numpy as np\n", - "import pandas as pd" - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
questionans
0小孩出生後應於何時申請育兒津貼?1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在...
1小孩出生後應於何時申請育兒津貼?隨時提出;津貼經審查通過後，追溯自受理申請之當月起發給。兒童出生後六十日內向戶政事務所完成出...
2育兒津貼申請應備文件為何?申請資料應備齊:(一)兒童之戶口名簿影本。(二)申請人之郵局存摺封面影本。(三)父母雙方身分...
3若民眾夫妻雙方均失業，是否可申請家庭育兒津貼費用補助一、育兒津貼補助對象：1.育有二足歲以下兒童。2.兒童之父母至少一方因育兒需要，致未能就業者...
4育兒津貼補助對象為何？育兒津貼補助對象，應符合下列規定：(一)育有二足歲以下兒童。(二)兒童之父母(或監護人)至少...
5如何認定是否符合父母未就業家庭育兒津貼中的未就業？1.未參加勞工「就業」保險（或軍保、公保）。2.最近一年度之稅捐稽徵機關核定之薪資所得及執行...
6未就業家庭育兒津貼條件中，若一方無工作但有投保公會勞保，請問是否可申請該項津貼，請提供該項業...針對勞工保險投保職業工會者（或雇主）等對象，一般均未參加「勞工就業保險」（自行加保者除外），...
7就業者家庭部分托育費用補助所稱的「就業者」為何？(1)受僱於政府、學校或公民營事業單位者。(2)勞動基準法所稱受雇主僱用從事工作者。(3)依...
8我未來想當合格托育人員，但又不是相關科系，新竹市哪裡會開辦托育人員訓練課程呢？勞動部勞動力發展署及本府委託社團法人新竹市社區婦女關懷協會、社團法人新竹市嬰幼兒保育協會、新...
9我該如何成為居家托育服務中心(原社區保母系統)的居家式托育人員呢？欲申請加入居家托育服務中心內的居家托育人員依規需年滿20歲並符合以下資格之一：1.取得保母技...
10親戚朋友托育，是否也可以領補助呢？一般民眾(含寶寶的爺爺奶奶、阿姨、叔叔、舅舅等三親等以內親屬)必須上課126小時托育人員訓練...
11就業者家庭部分托育費用補助申請應備文件為何？向誰申請1.補助申請表。2.最近戶籍謄本或戶口名簿影本。3.郵局存簿封面影本。4.托育契約書（親屬托...
12育兒津貼補助內容為為何？本津貼補助對象，應符合下列規定：(一)育有2足歲以下兒童。(二)兒童之父母(或監護人)至少一...
13誰可以提出育兒津貼申請手續？(一)申請人資格規定如下：1.育有二足歲以下兒童。2.兒童之父母(或監護人)至少一方因育兒需...
14托育費用補助標準為何？補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托...
15托育費用補助對象為何？補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總...
16育兒津貼補助金額多少？育兒津貼補助金額依家庭狀況而定，補助金額如下：(一)低收入戶：每名兒童每月補助新臺幣5000...
17請提供本年度保母系統(居家托育服務中心)相關資訊為何?目前本市居家托育服務中心承辦單位及聯絡電話如下：東區居家托育服務中心承辦單位：新竹市私立光復...
18育兒津貼受理單位為何？審核程序為何？本津貼申領及發放程序規定如下：(一)由申請人檢具相關證明文件向兒童戶籍地之區公所提出申請。(...
19要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服...
20就業者家庭部分托育費用補助申請對象及審核程序為何？就業者家庭部分托育費用補助的實施對象為：1.父母(或監護人)雙方或單親一方皆就業，或父母一方...
21就業者家庭部分托育費用補助審核程序為何？父母需於托育事實發生日起15日內，檢齊應備文件送交幼兒托育地點之社區保母系統或托嬰中心，其補...
22托育補助金額多少，補助期間多長？依各類家庭條件及送托托育人員資格，其托育補助金額說明如下：(一)將嬰幼兒送至托育服務中心中有...
23免費育兒指導的內容是什麼？政府建構友善的托育環境，除了提供托育費用及育兒津貼的經濟補助外，仍然重視親職教育，所以提供免...
24育兒指導員到府可提供哪些服務內容？（一）提供嬰幼兒進食、遊戲、睡眠及幼兒活動路線等適性與安全諮詢。（二）提供嬰幼兒穿衣、換尿布...
25目前有領取育嬰留職停薪不能請領育兒津貼，但育嬰留職停薪只能請領六個月，之後孩子仍未滿2歲，還...領取育嬰留職停薪津貼的六個月內，同一孩童不得再領取育兒津貼。但如果在育嬰留職停薪期間但未領取...
26居家式托育人員(原保母)或親屬保母每人至多可以照顧幾位幼童？居家式托育人員或親屬保母每人至多照顧兒童(含托育人員本人之幼兒)4人，其中未滿2歲者最多2人...
27申請育兒指導服務要符合怎樣的資格？只要符合下列資格之一，即可申請：（一）設籍且實際居住本市，家有0-2歲以下幼兒並有育兒指導需...
28如何申請育兒指導服務？（一）、電話諮詢且評估到府指導需求，服務電話：03-5396000。（二）、新竹市政府社會處...
29育兒指導服務可以提供哪些協助？全部免費服務嗎？育兒指導服務是新竹市政府社會處主辦的免費服務方案；提供電話親子教養諮詢，且評估實際需求後，由...
.........
297取用食物的時候要注意什麼嗎？106-12-31
298取用食物有限制數量及區域嗎？106-12-31
299放在冰箱中的食物安全嗎?106-12-31
300續食是什麼?106-12-31
301什麼是愛享冰箱?106-12-31
302推動社區愛享冰箱的目的?106-12-31
303請問如何進行合作社之清算？合作社（場）於奉令解散或陳准解散登記後，應即進行清算手續。合作社（場）之清算登記，旨在使得合...
304如何申請本市社會福利慈善基金會？業務範圍僅在新竹市者，向新竹市政府社會處申請，其設立基金最低數額為一千萬元以上。應備文件（各...
305本市人民團體申請設立程序及法令依據？依據人民團體法第8條規定人民團體之組織，應由發起人檢具申請書、章程草案及發起人名冊，向主管機...
306請問有關合作社變更登記應備文件為何？1.變更登記申請書。2.入社社員名冊。3.退（出）社社員名冊。4.年度社員大會紀錄(含改選理...
307本市人民團體選舉相關重要法令？人民團體選舉罷免辦法第5條人民團體理事、監事及會員代表之選舉或罷免，應由理事會在召開會議十五...
308請問合作社場解散之程序為何？合作社因左列各款情事之一而解散：1.章程所定解散之事由發生。2.社員大會之解散決議（應有全體...
309請問如何申請籌組合作社場？有共同需要實行合作的人，如要組織合作社（場）時，應有七人以上發起人，發起前應先洽詢主管機關索...
310申請籌組人民團體之條件為何？一、申請條件：須設籍於本市之市民30人以上發起二、發起人之基本資格：須本市市民、年滿二十歲，...
311何謂合作社？所謂合作社係指依平等原則，在互助組織之基礎上，以共同經營方法，謀社員經濟之利益與生活之改善，...
312如何申請設立財團法人社會福利慈善事業基金會？一、應檢附下列文件：1.申請書(一份)2.基金會概況表(一份)3.基金會籌備會議記錄(四份)...
313申請人民團體應附書表及證件為何？一、社會團體：1、新竹市社會團體申請書2、新竹市社會團體章程3、新竹市社會團體發起人名冊4、...
314如何加入志工行列？參考本府所編製之志願服務資源簡章，瞭解各類別志工隊服務內容及所需志工資格，再依本身志趣，逕向...
315如何申請志工隊依志願服務法所訂志願服務運用單位係指運用志工之機關、機構、學校、法人或經政府立案之團體。符合...
316如何申請志願服務紀錄冊需先完成12小時之志願服務基礎訓練和特殊訓練。完成上述訓練者，由志願服務運用單位志願服務運用...
317如何申請志願服務榮譽卡志工服務年資滿3年、服務時數達300小時以上者，得檢具1吋半身照2張、服務紀錄冊影本、申請表...
318加入國民年金保險，經審查符合所得未達一定標準者，如有戶籍遷徙，是否應重新申請？國民年金保費減免係依被保險人當月底戶籍所在地直轄市、縣（市）主管機關為準。故申請符合所得未達...
319申請國民年金所得未達一定標準資格認定，應向哪一單位提出申請？請攜帶申請人身分證、印章及全戶戶口名簿等資料至戶籍所在地區公所社政課提出申請。
320國民年金開辦時年滿65歲之一般國民，請領老年基本保證年金之資格為何？國民年金開辦時年滿65歲國民，在國內設有戶籍，且於最近3年內每年居住超過183日，而無下列情...
321國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於...
322國民年金之實施對象為何？1.年滿25歲、未滿65歲，在國內設有戶籍，且未參加勞保、農保、公教保、軍保之國民。2.國民...
323國民年金生育給付從什麼時候開辦？自100年7月1日實施。
324國民年金生育給付從何時開始調漲？自104年12月18日起調漲，由1個月的國民年金月投保金額調整至2個月。105年月投保金額為...
325國民年金生育給付的申請資格？1.生產期間需有國民年金保險身分。2.自100年7月1日後生產。3.生產日起算5年內需向各地...
326國民年金所得未達一定標準保費減免的申請方式？1.向戶籍所在地區公所申請及洽詢。2.攜帶身分證、印章、戶口名簿。3.全家人口之退休俸、學生...
\n", - "

326 rows × 2 columns

\n", - "
" - ], - "text/plain": [ - " question \\\n", - "0 小孩出生後應於何時申請育兒津貼? \n", - "1 小孩出生後應於何時申請育兒津貼? \n", - "2 育兒津貼申請應備文件為何? \n", - "3 若民眾夫妻雙方均失業，是否可申請家庭育兒津貼費用補助 \n", - "4 育兒津貼補助對象為何？ \n", - "5 如何認定是否符合父母未就業家庭育兒津貼中的未就業？ \n", - "6 未就業家庭育兒津貼條件中，若一方無工作但有投保公會勞保，請問是否可申請該項津貼，請提供該項業... \n", - "7 就業者家庭部分托育費用補助所稱的「就業者」為何？ \n", - "8 我未來想當合格托育人員，但又不是相關科系，新竹市哪裡會開辦托育人員訓練課程呢？ \n", - "9 我該如何成為居家托育服務中心(原社區保母系統)的居家式托育人員呢？ \n", - "10 親戚朋友托育，是否也可以領補助呢？ \n", - "11 就業者家庭部分托育費用補助申請應備文件為何？向誰申請 \n", - "12 育兒津貼補助內容為為何？ \n", - "13 誰可以提出育兒津貼申請手續？ \n", - "14 托育費用補助標準為何？ \n", - "15 托育費用補助對象為何？ \n", - "16 育兒津貼補助金額多少？ \n", - "17 請提供本年度保母系統(居家托育服務中心)相關資訊為何? \n", - "18 育兒津貼受理單位為何？審核程序為何？ \n", - "19 要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？ \n", - "20 就業者家庭部分托育費用補助申請對象及審核程序為何？ \n", - "21 就業者家庭部分托育費用補助審核程序為何？ \n", - "22 托育補助金額多少，補助期間多長？ \n", - "23 免費育兒指導的內容是什麼？ \n", - "24 育兒指導員到府可提供哪些服務內容？ \n", - "25 目前有領取育嬰留職停薪不能請領育兒津貼，但育嬰留職停薪只能請領六個月，之後孩子仍未滿2歲，還... \n", - "26 居家式托育人員(原保母)或親屬保母每人至多可以照顧幾位幼童？ \n", - "27 申請育兒指導服務要符合怎樣的資格？ \n", - "28 如何申請育兒指導服務？ \n", - "29 育兒指導服務可以提供哪些協助？全部免費服務嗎？ \n", - ".. ... \n", - "297 取用食物的時候要注意什麼嗎？ \n", - "298 取用食物有限制數量及區域嗎？ \n", - "299 放在冰箱中的食物安全嗎? \n", - "300 續食是什麼? \n", - "301 什麼是愛享冰箱? \n", - "302 推動社區愛享冰箱的目的? \n", - "303 請問如何進行合作社之清算？ \n", - "304 如何申請本市社會福利慈善基金會？ \n", - "305 本市人民團體申請設立程序及法令依據？ \n", - "306 請問有關合作社變更登記應備文件為何？ \n", - "307 本市人民團體選舉相關重要法令？ \n", - "308 請問合作社場解散之程序為何？ \n", - "309 請問如何申請籌組合作社場？ \n", - "310 申請籌組人民團體之條件為何？ \n", - "311 何謂合作社？ \n", - "312 如何申請設立財團法人社會福利慈善事業基金會？ \n", - "313 申請人民團體應附書表及證件為何？ \n", - "314 如何加入志工行列？ \n", - "315 如何申請志工隊 \n", - "316 如何申請志願服務紀錄冊 \n", - "317 如何申請志願服務榮譽卡 \n", - "318 加入國民年金保險，經審查符合所得未達一定標準者，如有戶籍遷徙，是否應重新申請？ \n", - "319 申請國民年金所得未達一定標準資格認定，應向哪一單位提出申請？ \n", - "320 國民年金開辦時年滿65歲之一般國民，請領老年基本保證年金之資格為何？ \n", - "321 國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？ \n", - "322 國民年金之實施對象為何？ \n", - "323 國民年金生育給付從什麼時候開辦？ \n", - "324 國民年金生育給付從何時開始調漲？ \n", - "325 國民年金生育給付的申請資格？ \n", - "326 國民年金所得未達一定標準保費減免的申請方式？ \n", - "\n", - " ans \n", - "0 1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在... \n", - "1 隨時提出;津貼經審查通過後，追溯自受理申請之當月起發給。兒童出生後六十日內向戶政事務所完成出... \n", - "2 申請資料應備齊:(一)兒童之戶口名簿影本。(二)申請人之郵局存摺封面影本。(三)父母雙方身分... \n", - "3 一、育兒津貼補助對象：1.育有二足歲以下兒童。2.兒童之父母至少一方因育兒需要，致未能就業者... \n", - "4 育兒津貼補助對象，應符合下列規定：(一)育有二足歲以下兒童。(二)兒童之父母(或監護人)至少... \n", - "5 1.未參加勞工「就業」保險（或軍保、公保）。2.最近一年度之稅捐稽徵機關核定之薪資所得及執行... \n", - "6 針對勞工保險投保職業工會者（或雇主）等對象，一般均未參加「勞工就業保險」（自行加保者除外），... \n", - "7 (1)受僱於政府、學校或公民營事業單位者。(2)勞動基準法所稱受雇主僱用從事工作者。(3)依... \n", - "8 勞動部勞動力發展署及本府委託社團法人新竹市社區婦女關懷協會、社團法人新竹市嬰幼兒保育協會、新... \n", - "9 欲申請加入居家托育服務中心內的居家托育人員依規需年滿20歲並符合以下資格之一：1.取得保母技... \n", - "10 一般民眾(含寶寶的爺爺奶奶、阿姨、叔叔、舅舅等三親等以內親屬)必須上課126小時托育人員訓練... \n", - "11 1.補助申請表。2.最近戶籍謄本或戶口名簿影本。3.郵局存簿封面影本。4.托育契約書（親屬托... \n", - "12 本津貼補助對象，應符合下列規定：(一)育有2足歲以下兒童。(二)兒童之父母(或監護人)至少一... \n", - "13 (一)申請人資格規定如下：1.育有二足歲以下兒童。2.兒童之父母(或監護人)至少一方因育兒需... \n", - "14 補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托... \n", - "15 補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總... \n", - "16 育兒津貼補助金額依家庭狀況而定，補助金額如下：(一)低收入戶：每名兒童每月補助新臺幣5000... \n", - "17 目前本市居家托育服務中心承辦單位及聯絡電話如下：東區居家托育服務中心承辦單位：新竹市私立光復... \n", - "18 本津貼申領及發放程序規定如下：(一)由申請人檢具相關證明文件向兒童戶籍地之區公所提出申請。(... \n", - "19 依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服... \n", - "20 就業者家庭部分托育費用補助的實施對象為：1.父母(或監護人)雙方或單親一方皆就業，或父母一方... \n", - "21 父母需於托育事實發生日起15日內，檢齊應備文件送交幼兒托育地點之社區保母系統或托嬰中心，其補... \n", - "22 依各類家庭條件及送托托育人員資格，其托育補助金額說明如下：(一)將嬰幼兒送至托育服務中心中有... \n", - "23 政府建構友善的托育環境，除了提供托育費用及育兒津貼的經濟補助外，仍然重視親職教育，所以提供免... \n", - "24 （一）提供嬰幼兒進食、遊戲、睡眠及幼兒活動路線等適性與安全諮詢。（二）提供嬰幼兒穿衣、換尿布... \n", - "25 領取育嬰留職停薪津貼的六個月內，同一孩童不得再領取育兒津貼。但如果在育嬰留職停薪期間但未領取... \n", - "26 居家式托育人員或親屬保母每人至多照顧兒童(含托育人員本人之幼兒)4人，其中未滿2歲者最多2人... \n", - "27 只要符合下列資格之一，即可申請：（一）設籍且實際居住本市，家有0-2歲以下幼兒並有育兒指導需... \n", - "28 （一）、電話諮詢且評估到府指導需求，服務電話：03-5396000。（二）、新竹市政府社會處... \n", - "29 育兒指導服務是新竹市政府社會處主辦的免費服務方案；提供電話親子教養諮詢，且評估實際需求後，由... \n", - ".. ... \n", - "297 106-12-31 \n", - "298 106-12-31 \n", - "299 106-12-31 \n", - "300 106-12-31 \n", - "301 106-12-31 \n", - "302 106-12-31 \n", - "303 合作社（場）於奉令解散或陳准解散登記後，應即進行清算手續。合作社（場）之清算登記，旨在使得合... \n", - "304 業務範圍僅在新竹市者，向新竹市政府社會處申請，其設立基金最低數額為一千萬元以上。應備文件（各... \n", - "305 依據人民團體法第8條規定人民團體之組織，應由發起人檢具申請書、章程草案及發起人名冊，向主管機... \n", - "306 1.變更登記申請書。2.入社社員名冊。3.退（出）社社員名冊。4.年度社員大會紀錄(含改選理... \n", - "307 人民團體選舉罷免辦法第5條人民團體理事、監事及會員代表之選舉或罷免，應由理事會在召開會議十五... \n", - "308 合作社因左列各款情事之一而解散：1.章程所定解散之事由發生。2.社員大會之解散決議（應有全體... \n", - "309 有共同需要實行合作的人，如要組織合作社（場）時，應有七人以上發起人，發起前應先洽詢主管機關索... \n", - "310 一、申請條件：須設籍於本市之市民30人以上發起二、發起人之基本資格：須本市市民、年滿二十歲，... \n", - "311 所謂合作社係指依平等原則，在互助組織之基礎上，以共同經營方法，謀社員經濟之利益與生活之改善，... \n", - "312 一、應檢附下列文件：1.申請書(一份)2.基金會概況表(一份)3.基金會籌備會議記錄(四份)... \n", - "313 一、社會團體：1、新竹市社會團體申請書2、新竹市社會團體章程3、新竹市社會團體發起人名冊4、... \n", - "314 參考本府所編製之志願服務資源簡章，瞭解各類別志工隊服務內容及所需志工資格，再依本身志趣，逕向... \n", - "315 依志願服務法所訂志願服務運用單位係指運用志工之機關、機構、學校、法人或經政府立案之團體。符合... \n", - "316 需先完成12小時之志願服務基礎訓練和特殊訓練。完成上述訓練者，由志願服務運用單位志願服務運用... \n", - "317 志工服務年資滿3年、服務時數達300小時以上者，得檢具1吋半身照2張、服務紀錄冊影本、申請表... \n", - "318 國民年金保費減免係依被保險人當月底戶籍所在地直轄市、縣（市）主管機關為準。故申請符合所得未達... \n", - "319 請攜帶申請人身分證、印章及全戶戶口名簿等資料至戶籍所在地區公所社政課提出申請。 \n", - "320 國民年金開辦時年滿65歲國民，在國內設有戶籍，且於最近3年內每年居住超過183日，而無下列情... \n", - "321 國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於... \n", - "322 1.年滿25歲、未滿65歲，在國內設有戶籍，且未參加勞保、農保、公教保、軍保之國民。2.國民... \n", - "323 自100年7月1日實施。 \n", - "324 自104年12月18日起調漲，由1個月的國民年金月投保金額調整至2個月。105年月投保金額為... \n", - "325 1.生產期間需有國民年金保險身分。2.自100年7月1日後生產。3.生產日起算5年內需向各地... \n", - "326 1.向戶籍所在地區公所申請及洽詢。2.攜帶身分證、印章、戶口名簿。3.全家人口之退休俸、學生... \n", - "\n", - "[326 rows x 2 columns]" - ] - }, - "execution_count": 2, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "# 把檔案讀出來\n", - "df_QA = pd.read_json('ProcessedData.json', encoding='utf8')\n", - "# 我們這次只會使用到question跟ans這兩個欄位\n", - "df_question = df_QA[['question', 'ans']].copy() ## 不要更動到原始的DataFrame\n", - "df_question.drop_duplicates(inplace=True) ## 丟掉重複的資料\n", - "df_question ## show出來" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": { - "scrolled": true - }, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "Building prefix dict from D:\\Projects\\IIIMaterial\\09_IrImplementation\\dict.txt.big ...\n", - "Dumping model to file cache C:\\Users\\user\\AppData\\Local\\Temp\\jieba.u833a2d8ba23239541503666f62a3e4bd.cache\n", - "Loading model cost 3.481 seconds.\n", - "Prefix dict has been built succesfully.\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
questionansprocessed
0小孩出生後應於何時申請育兒津貼?1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在...[小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ]
1小孩出生後應於何時申請育兒津貼?隨時提出;津貼經審查通過後，追溯自受理申請之當月起發給。兒童出生後六十日內向戶政事務所完成出...[小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ]
2育兒津貼申請應備文件為何?申請資料應備齊:(一)兒童之戶口名簿影本。(二)申請人之郵局存摺封面影本。(三)父母雙方身分...[育兒, 津貼, 申請, 應, 備, 文件, 為, 何, , ]
3若民眾夫妻雙方均失業，是否可申請家庭育兒津貼費用補助一、育兒津貼補助對象：1.育有二足歲以下兒童。2.兒童之父母至少一方因育兒需要，致未能就業者...[若, 民, 眾, 夫妻, 雙方, 均, 失業, , , 是否, 可, 申請, 家庭, 育兒...
4育兒津貼補助對象為何？育兒津貼補助對象，應符合下列規定：(一)育有二足歲以下兒童。(二)兒童之父母(或監護人)至少...[育兒, 津貼, 貼補, 補助, 對象, 為, 何, , ]
5如何認定是否符合父母未就業家庭育兒津貼中的未就業？1.未參加勞工「就業」保險（或軍保、公保）。2.最近一年度之稅捐稽徵機關核定之薪資所得及執行...[如何, 認定, 是否, 符合, 父母, 未, 就業, 家庭, 育兒, 津貼, 中, 的, ...
6未就業家庭育兒津貼條件中，若一方無工作但有投保公會勞保，請問是否可申請該項津貼，請提供該項業...針對勞工保險投保職業工會者（或雇主）等對象，一般均未參加「勞工就業保險」（自行加保者除外），...[未, 就業, 家庭, 育兒, 津貼, 條件, 中, , , 若, 一方, 無, 工作, 但...
7就業者家庭部分托育費用補助所稱的「就業者」為何？(1)受僱於政府、學校或公民營事業單位者。(2)勞動基準法所稱受雇主僱用從事工作者。(3)依...[就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 所稱, 的, , ,...
8我未來想當合格托育人員，但又不是相關科系，新竹市哪裡會開辦托育人員訓練課程呢？勞動部勞動力發展署及本府委託社團法人新竹市社區婦女關懷協會、社團法人新竹市嬰幼兒保育協會、新...[我, 未來, 想, 當, 合格, 托, 育人, 人員, , , 但, 又, 不是, 相關,...
9我該如何成為居家托育服務中心(原社區保母系統)的居家式托育人員呢？欲申請加入居家托育服務中心內的居家托育人員依規需年滿20歲並符合以下資格之一：1.取得保母技...[我, 該, 如何, 成, 為, 居家, 托, 育, 服務, 服務中心, 中心, , , 原...
10親戚朋友托育，是否也可以領補助呢？一般民眾(含寶寶的爺爺奶奶、阿姨、叔叔、舅舅等三親等以內親屬)必須上課126小時托育人員訓練...[親戚, 親戚朋友, 朋友, 托, 育, , , 是否, 也, 可以, 領, 補助, 呢, , ]
11就業者家庭部分托育費用補助申請應備文件為何？向誰申請1.補助申請表。2.最近戶籍謄本或戶口名簿影本。3.郵局存簿封面影本。4.托育契約書（親屬托...[就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 申請, 應, 備, ...
12育兒津貼補助內容為為何？本津貼補助對象，應符合下列規定：(一)育有2足歲以下兒童。(二)兒童之父母(或監護人)至少一...[育兒, 津貼, 貼補, 補助, 內容, 為, 為, 何, , ]
13誰可以提出育兒津貼申請手續？(一)申請人資格規定如下：1.育有二足歲以下兒童。2.兒童之父母(或監護人)至少一方因育兒需...[誰, 可以, 提出, 育兒, 津貼, 申請, 手續, , ]
14托育費用補助標準為何？補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托...[托, 育, 費用, 補助, 標準, 為, 何, , ]
15托育費用補助對象為何？補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總...[托, 育, 費用, 補助, 對象, 為, 何, , ]
16育兒津貼補助金額多少？育兒津貼補助金額依家庭狀況而定，補助金額如下：(一)低收入戶：每名兒童每月補助新臺幣5000...[育兒, 津貼, 貼補, 補助, 補助金, 金額, 多少, , ]
17請提供本年度保母系統(居家托育服務中心)相關資訊為何?目前本市居家托育服務中心承辦單位及聯絡電話如下：東區居家托育服務中心承辦單位：新竹市私立光復...[請, 提供, 本年, 本年度, 年度, 保, 母系, 系統, , , 居家, 托, 育, ...
18育兒津貼受理單位為何？審核程序為何？本津貼申領及發放程序規定如下：(一)由申請人檢具相關證明文件向兒童戶籍地之區公所提出申請。(...[育兒, 津貼, 受理, 單位, 為, 何, , , 審, 核, 程序, 為, 何, , ]
19要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服...[要, 申請, 托, 育, 補助, 一定, 定要, 找, 加入, 居家, 托, 育, 服務,...
20就業者家庭部分托育費用補助申請對象及審核程序為何？就業者家庭部分托育費用補助的實施對象為：1.父母(或監護人)雙方或單親一方皆就業，或父母一方...[就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 申請, 對象, 及,...
21就業者家庭部分托育費用補助審核程序為何？父母需於托育事實發生日起15日內，檢齊應備文件送交幼兒托育地點之社區保母系統或托嬰中心，其補...[就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 審, 核, 程序, ...
22托育補助金額多少，補助期間多長？依各類家庭條件及送托托育人員資格，其托育補助金額說明如下：(一)將嬰幼兒送至托育服務中心中有...[托, 育, 補助, 補助金, 金額, 多少, , , 補助, 期間, 多長, , ]
23免費育兒指導的內容是什麼？政府建構友善的托育環境，除了提供托育費用及育兒津貼的經濟補助外，仍然重視親職教育，所以提供免...[免費, 育兒, 指導, 的, 內容, 是, 什麼, , ]
24育兒指導員到府可提供哪些服務內容？（一）提供嬰幼兒進食、遊戲、睡眠及幼兒活動路線等適性與安全諮詢。（二）提供嬰幼兒穿衣、換尿布...[育兒, 指導, 指導員, 到, 府, 可, 提供, 哪些, 服務, 內容, , , ]
25目前有領取育嬰留職停薪不能請領育兒津貼，但育嬰留職停薪只能請領六個月，之後孩子仍未滿2歲，還...領取育嬰留職停薪津貼的六個月內，同一孩童不得再領取育兒津貼。但如果在育嬰留職停薪期間但未領取...[目前, 有, 領取, 育嬰, 留職, 留職停薪, 停薪, 不能, 請領, 育兒, 津貼, ...
26居家式托育人員(原保母)或親屬保母每人至多可以照顧幾位幼童？居家式托育人員或親屬保母每人至多照顧兒童(含托育人員本人之幼兒)4人，其中未滿2歲者最多2人...[居家, 式, 托, 育人, 人員, , , 原, 保, 母, , , 或, 親屬, 保, ...
27申請育兒指導服務要符合怎樣的資格？只要符合下列資格之一，即可申請：（一）設籍且實際居住本市，家有0-2歲以下幼兒並有育兒指導需...[申請, 育兒, 指導, 服務, 務要, 符合, 怎樣, 的, 資格, , ]
28如何申請育兒指導服務？（一）、電話諮詢且評估到府指導需求，服務電話：03-5396000。（二）、新竹市政府社會處...[如何, 申請, 育兒, 指導, 服務, , ]
29育兒指導服務可以提供哪些協助？全部免費服務嗎？育兒指導服務是新竹市政府社會處主辦的免費服務方案；提供電話親子教養諮詢，且評估實際需求後，由...[育兒, 指導, 服務, 可以, 提供, 哪些, 協助, , , 全部, 免費, 服務, 嗎...
............
297取用食物的時候要注意什麼嗎？106-12-31[取用, 食物, 的, 時候, 要, 注意, 什麼, 嗎, , ]
298取用食物有限制數量及區域嗎？106-12-31[取用, 食物, 有限, 限制, 數量, 及, 區域, 嗎, , ]
299放在冰箱中的食物安全嗎?106-12-31[放在, 冰箱, 中, 的, 食物, 安全, 嗎, , ]
300續食是什麼?106-12-31[續, 食, 是, 什麼, , ]
301什麼是愛享冰箱?106-12-31[什麼, 是, 愛, 享, 冰箱, , ]
302推動社區愛享冰箱的目的?106-12-31[推動, 社區, 愛, 享, 冰箱, 的, 目的, , ]
303請問如何進行合作社之清算？合作社（場）於奉令解散或陳准解散登記後，應即進行清算手續。合作社（場）之清算登記，旨在使得合...[請問, 如何, 進行, 合作, 合作社, 之, 清算, , ]
304如何申請本市社會福利慈善基金會？業務範圍僅在新竹市者，向新竹市政府社會處申請，其設立基金最低數額為一千萬元以上。應備文件（各...[如何, 申請, 本市, 社會, 社會福利, 福利, 慈善, 基金, 基金會, , ]
305本市人民團體申請設立程序及法令依據？依據人民團體法第8條規定人民團體之組織，應由發起人檢具申請書、章程草案及發起人名冊，向主管機...[本市, 人民, 人民團體, 民團, 團體, 申請, 設立, 程序, 及, 法令, 依據, , ]
306請問有關合作社變更登記應備文件為何？1.變更登記申請書。2.入社社員名冊。3.退（出）社社員名冊。4.年度社員大會紀錄(含改選理...[請問, 有關, 關合, 合作, 合作社, 變更, 登記, 應, 備, 文件, 為, 何, , ]
307本市人民團體選舉相關重要法令？人民團體選舉罷免辦法第5條人民團體理事、監事及會員代表之選舉或罷免，應由理事會在召開會議十五...[本市, 人民, 人民團體, 民團, 團體, 選舉, 相關, 重要, 法令, , ]
308請問合作社場解散之程序為何？合作社因左列各款情事之一而解散：1.章程所定解散之事由發生。2.社員大會之解散決議（應有全體...[請問, 合作, 合作社, 場, 解散, 之, 程序, 為, 何, , ]
309請問如何申請籌組合作社場？有共同需要實行合作的人，如要組織合作社（場）時，應有七人以上發起人，發起前應先洽詢主管機關索...[請問, 如何, 申請, 籌組, 組合, 合作, 合作社, 場, , ]
310申請籌組人民團體之條件為何？一、申請條件：須設籍於本市之市民30人以上發起二、發起人之基本資格：須本市市民、年滿二十歲，...[申請, 籌組, 組人, 人民, 人民團體, 民團, 團體, 之, 條件, 為, 何, , ]
311何謂合作社？所謂合作社係指依平等原則，在互助組織之基礎上，以共同經營方法，謀社員經濟之利益與生活之改善，...[何謂, 合作, 合作社, , ]
312如何申請設立財團法人社會福利慈善事業基金會？一、應檢附下列文件：1.申請書(一份)2.基金會概況表(一份)3.基金會籌備會議記錄(四份)...[如何, 申請, 設立, 財團, 財團法人, 法人, 社會, 社會福利, 福利, 慈善, 慈...
313申請人民團體應附書表及證件為何？一、社會團體：1、新竹市社會團體申請書2、新竹市社會團體章程3、新竹市社會團體發起人名冊4、...[申請, 申請人, 人民, 人民團體, 民團, 團體, 應, 附, 書, 表, 及, 證件,...
314如何加入志工行列？參考本府所編製之志願服務資源簡章，瞭解各類別志工隊服務內容及所需志工資格，再依本身志趣，逕向...[如何, 加入, 志, 工行, 行列, , ]
315如何申請志工隊依志願服務法所訂志願服務運用單位係指運用志工之機關、機構、學校、法人或經政府立案之團體。符合...[如何, 申請, 志, 工, 隊]
316如何申請志願服務紀錄冊需先完成12小時之志願服務基礎訓練和特殊訓練。完成上述訓練者，由志願服務運用單位志願服務運用...[如何, 申請, 志願, 服務, 紀錄, 冊]
317如何申請志願服務榮譽卡志工服務年資滿3年、服務時數達300小時以上者，得檢具1吋半身照2張、服務紀錄冊影本、申請表...[如何, 申請, 志願, 服務, 榮譽, 卡]
318加入國民年金保險，經審查符合所得未達一定標準者，如有戶籍遷徙，是否應重新申請？國民年金保費減免係依被保險人當月底戶籍所在地直轄市、縣（市）主管機關為準。故申請符合所得未達...[加入, 國民, 年金, 保險, , , 經審查, 審查, 符合, 所得, 未, 達, 一定...
319申請國民年金所得未達一定標準資格認定，應向哪一單位提出申請？請攜帶申請人身分證、印章及全戶戶口名簿等資料至戶籍所在地區公所社政課提出申請。[申請, 國民, 年金, 所得, 未, 達, 一定, 定標, 標準, 資格, 認定, , ,...
320國民年金開辦時年滿65歲之一般國民，請領老年基本保證年金之資格為何？國民年金開辦時年滿65歲國民，在國內設有戶籍，且於最近3年內每年居住超過183日，而無下列情...[國民, 年金, 開辦, 時, 年滿, 65, 歲, 之一, 一般, 國民, , , 請領,...
321國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於...[國民, 年金, 保險, 被保險人, 保險, 保險人, 如果, 是, 家庭, 家庭收入, 收...
322國民年金之實施對象為何？1.年滿25歲、未滿65歲，在國內設有戶籍，且未參加勞保、農保、公教保、軍保之國民。2.國民...[國民, 年金, 之, 實施, 對象, 為, 何, , ]
323國民年金生育給付從什麼時候開辦？自100年7月1日實施。[國民, 年金, 金生, 生育, 給付, 從, 什麼, 時候, 開辦, , ]
324國民年金生育給付從何時開始調漲？自104年12月18日起調漲，由1個月的國民年金月投保金額調整至2個月。105年月投保金額為...[國民, 年金, 金生, 生育, 給付, 從, 何時, 開始, 調, 漲, , ]
325國民年金生育給付的申請資格？1.生產期間需有國民年金保險身分。2.自100年7月1日後生產。3.生產日起算5年內需向各地...[國民, 年金, 金生, 生育, 給付, 的, 申請, 資格, , ]
326國民年金所得未達一定標準保費減免的申請方式？1.向戶籍所在地區公所申請及洽詢。2.攜帶身分證、印章、戶口名簿。3.全家人口之退休俸、學生...[國民, 年金, 所得, 未, 達, 一定, 定標, 標準, 準保, 保費, 減免, 的, ...
\n", - "

326 rows × 3 columns

\n", - "
" - ], - "text/plain": [ - " question \\\n", - "0 小孩出生後應於何時申請育兒津貼? \n", - "1 小孩出生後應於何時申請育兒津貼? \n", - "2 育兒津貼申請應備文件為何? \n", - "3 若民眾夫妻雙方均失業，是否可申請家庭育兒津貼費用補助 \n", - "4 育兒津貼補助對象為何？ \n", - "5 如何認定是否符合父母未就業家庭育兒津貼中的未就業？ \n", - "6 未就業家庭育兒津貼條件中，若一方無工作但有投保公會勞保，請問是否可申請該項津貼，請提供該項業... \n", - "7 就業者家庭部分托育費用補助所稱的「就業者」為何？ \n", - "8 我未來想當合格托育人員，但又不是相關科系，新竹市哪裡會開辦托育人員訓練課程呢？ \n", - "9 我該如何成為居家托育服務中心(原社區保母系統)的居家式托育人員呢？ \n", - "10 親戚朋友托育，是否也可以領補助呢？ \n", - "11 就業者家庭部分托育費用補助申請應備文件為何？向誰申請 \n", - "12 育兒津貼補助內容為為何？ \n", - "13 誰可以提出育兒津貼申請手續？ \n", - "14 托育費用補助標準為何？ \n", - "15 托育費用補助對象為何？ \n", - "16 育兒津貼補助金額多少？ \n", - "17 請提供本年度保母系統(居家托育服務中心)相關資訊為何? \n", - "18 育兒津貼受理單位為何？審核程序為何？ \n", - "19 要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？ \n", - "20 就業者家庭部分托育費用補助申請對象及審核程序為何？ \n", - "21 就業者家庭部分托育費用補助審核程序為何？ \n", - "22 托育補助金額多少，補助期間多長？ \n", - "23 免費育兒指導的內容是什麼？ \n", - "24 育兒指導員到府可提供哪些服務內容？ \n", - "25 目前有領取育嬰留職停薪不能請領育兒津貼，但育嬰留職停薪只能請領六個月，之後孩子仍未滿2歲，還... \n", - "26 居家式托育人員(原保母)或親屬保母每人至多可以照顧幾位幼童？ \n", - "27 申請育兒指導服務要符合怎樣的資格？ \n", - "28 如何申請育兒指導服務？ \n", - "29 育兒指導服務可以提供哪些協助？全部免費服務嗎？ \n", - ".. ... \n", - "297 取用食物的時候要注意什麼嗎？ \n", - "298 取用食物有限制數量及區域嗎？ \n", - "299 放在冰箱中的食物安全嗎? \n", - "300 續食是什麼? \n", - "301 什麼是愛享冰箱? \n", - "302 推動社區愛享冰箱的目的? \n", - "303 請問如何進行合作社之清算？ \n", - "304 如何申請本市社會福利慈善基金會？ \n", - "305 本市人民團體申請設立程序及法令依據？ \n", - "306 請問有關合作社變更登記應備文件為何？ \n", - "307 本市人民團體選舉相關重要法令？ \n", - "308 請問合作社場解散之程序為何？ \n", - "309 請問如何申請籌組合作社場？ \n", - "310 申請籌組人民團體之條件為何？ \n", - "311 何謂合作社？ \n", - "312 如何申請設立財團法人社會福利慈善事業基金會？ \n", - "313 申請人民團體應附書表及證件為何？ \n", - "314 如何加入志工行列？ \n", - "315 如何申請志工隊 \n", - "316 如何申請志願服務紀錄冊 \n", - "317 如何申請志願服務榮譽卡 \n", - "318 加入國民年金保險，經審查符合所得未達一定標準者，如有戶籍遷徙，是否應重新申請？ \n", - "319 申請國民年金所得未達一定標準資格認定，應向哪一單位提出申請？ \n", - "320 國民年金開辦時年滿65歲之一般國民，請領老年基本保證年金之資格為何？ \n", - "321 國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？ \n", - "322 國民年金之實施對象為何？ \n", - "323 國民年金生育給付從什麼時候開辦？ \n", - "324 國民年金生育給付從何時開始調漲？ \n", - "325 國民年金生育給付的申請資格？ \n", - "326 國民年金所得未達一定標準保費減免的申請方式？ \n", - "\n", - " ans \\\n", - "0 1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在... \n", - "1 隨時提出;津貼經審查通過後，追溯自受理申請之當月起發給。兒童出生後六十日內向戶政事務所完成出... \n", - "2 申請資料應備齊:(一)兒童之戶口名簿影本。(二)申請人之郵局存摺封面影本。(三)父母雙方身分... \n", - "3 一、育兒津貼補助對象：1.育有二足歲以下兒童。2.兒童之父母至少一方因育兒需要，致未能就業者... \n", - "4 育兒津貼補助對象，應符合下列規定：(一)育有二足歲以下兒童。(二)兒童之父母(或監護人)至少... \n", - "5 1.未參加勞工「就業」保險（或軍保、公保）。2.最近一年度之稅捐稽徵機關核定之薪資所得及執行... \n", - "6 針對勞工保險投保職業工會者（或雇主）等對象，一般均未參加「勞工就業保險」（自行加保者除外），... \n", - "7 (1)受僱於政府、學校或公民營事業單位者。(2)勞動基準法所稱受雇主僱用從事工作者。(3)依... \n", - "8 勞動部勞動力發展署及本府委託社團法人新竹市社區婦女關懷協會、社團法人新竹市嬰幼兒保育協會、新... \n", - "9 欲申請加入居家托育服務中心內的居家托育人員依規需年滿20歲並符合以下資格之一：1.取得保母技... \n", - "10 一般民眾(含寶寶的爺爺奶奶、阿姨、叔叔、舅舅等三親等以內親屬)必須上課126小時托育人員訓練... \n", - "11 1.補助申請表。2.最近戶籍謄本或戶口名簿影本。3.郵局存簿封面影本。4.托育契約書（親屬托... \n", - "12 本津貼補助對象，應符合下列規定：(一)育有2足歲以下兒童。(二)兒童之父母(或監護人)至少一... \n", - "13 (一)申請人資格規定如下：1.育有二足歲以下兒童。2.兒童之父母(或監護人)至少一方因育兒需... \n", - "14 補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托... \n", - "15 補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總... \n", - "16 育兒津貼補助金額依家庭狀況而定，補助金額如下：(一)低收入戶：每名兒童每月補助新臺幣5000... \n", - "17 目前本市居家托育服務中心承辦單位及聯絡電話如下：東區居家托育服務中心承辦單位：新竹市私立光復... \n", - "18 本津貼申領及發放程序規定如下：(一)由申請人檢具相關證明文件向兒童戶籍地之區公所提出申請。(... \n", - "19 依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服... \n", - "20 就業者家庭部分托育費用補助的實施對象為：1.父母(或監護人)雙方或單親一方皆就業，或父母一方... \n", - "21 父母需於托育事實發生日起15日內，檢齊應備文件送交幼兒托育地點之社區保母系統或托嬰中心，其補... \n", - "22 依各類家庭條件及送托托育人員資格，其托育補助金額說明如下：(一)將嬰幼兒送至托育服務中心中有... \n", - "23 政府建構友善的托育環境，除了提供托育費用及育兒津貼的經濟補助外，仍然重視親職教育，所以提供免... \n", - "24 （一）提供嬰幼兒進食、遊戲、睡眠及幼兒活動路線等適性與安全諮詢。（二）提供嬰幼兒穿衣、換尿布... \n", - "25 領取育嬰留職停薪津貼的六個月內，同一孩童不得再領取育兒津貼。但如果在育嬰留職停薪期間但未領取... \n", - "26 居家式托育人員或親屬保母每人至多照顧兒童(含托育人員本人之幼兒)4人，其中未滿2歲者最多2人... \n", - "27 只要符合下列資格之一，即可申請：（一）設籍且實際居住本市，家有0-2歲以下幼兒並有育兒指導需... \n", - "28 （一）、電話諮詢且評估到府指導需求，服務電話：03-5396000。（二）、新竹市政府社會處... \n", - "29 育兒指導服務是新竹市政府社會處主辦的免費服務方案；提供電話親子教養諮詢，且評估實際需求後，由... \n", - ".. ... \n", - "297 106-12-31 \n", - "298 106-12-31 \n", - "299 106-12-31 \n", - "300 106-12-31 \n", - "301 106-12-31 \n", - "302 106-12-31 \n", - "303 合作社（場）於奉令解散或陳准解散登記後，應即進行清算手續。合作社（場）之清算登記，旨在使得合... \n", - "304 業務範圍僅在新竹市者，向新竹市政府社會處申請，其設立基金最低數額為一千萬元以上。應備文件（各... \n", - "305 依據人民團體法第8條規定人民團體之組織，應由發起人檢具申請書、章程草案及發起人名冊，向主管機... \n", - "306 1.變更登記申請書。2.入社社員名冊。3.退（出）社社員名冊。4.年度社員大會紀錄(含改選理... \n", - "307 人民團體選舉罷免辦法第5條人民團體理事、監事及會員代表之選舉或罷免，應由理事會在召開會議十五... \n", - "308 合作社因左列各款情事之一而解散：1.章程所定解散之事由發生。2.社員大會之解散決議（應有全體... \n", - "309 有共同需要實行合作的人，如要組織合作社（場）時，應有七人以上發起人，發起前應先洽詢主管機關索... \n", - "310 一、申請條件：須設籍於本市之市民30人以上發起二、發起人之基本資格：須本市市民、年滿二十歲，... \n", - "311 所謂合作社係指依平等原則，在互助組織之基礎上，以共同經營方法，謀社員經濟之利益與生活之改善，... \n", - "312 一、應檢附下列文件：1.申請書(一份)2.基金會概況表(一份)3.基金會籌備會議記錄(四份)... \n", - "313 一、社會團體：1、新竹市社會團體申請書2、新竹市社會團體章程3、新竹市社會團體發起人名冊4、... \n", - "314 參考本府所編製之志願服務資源簡章，瞭解各類別志工隊服務內容及所需志工資格，再依本身志趣，逕向... \n", - "315 依志願服務法所訂志願服務運用單位係指運用志工之機關、機構、學校、法人或經政府立案之團體。符合... \n", - "316 需先完成12小時之志願服務基礎訓練和特殊訓練。完成上述訓練者，由志願服務運用單位志願服務運用... \n", - "317 志工服務年資滿3年、服務時數達300小時以上者，得檢具1吋半身照2張、服務紀錄冊影本、申請表... \n", - "318 國民年金保費減免係依被保險人當月底戶籍所在地直轄市、縣（市）主管機關為準。故申請符合所得未達... \n", - "319 請攜帶申請人身分證、印章及全戶戶口名簿等資料至戶籍所在地區公所社政課提出申請。 \n", - "320 國民年金開辦時年滿65歲國民，在國內設有戶籍，且於最近3年內每年居住超過183日，而無下列情... \n", - "321 國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於... \n", - "322 1.年滿25歲、未滿65歲，在國內設有戶籍，且未參加勞保、農保、公教保、軍保之國民。2.國民... \n", - "323 自100年7月1日實施。 \n", - "324 自104年12月18日起調漲，由1個月的國民年金月投保金額調整至2個月。105年月投保金額為... \n", - "325 1.生產期間需有國民年金保險身分。2.自100年7月1日後生產。3.生產日起算5年內需向各地... \n", - "326 1.向戶籍所在地區公所申請及洽詢。2.攜帶身分證、印章、戶口名簿。3.全家人口之退休俸、學生... \n", - "\n", - " processed \n", - "0 [小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ] \n", - "1 [小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ] \n", - "2 [育兒, 津貼, 申請, 應, 備, 文件, 為, 何, , ] \n", - "3 [若, 民, 眾, 夫妻, 雙方, 均, 失業, , , 是否, 可, 申請, 家庭, 育兒... \n", - "4 [育兒, 津貼, 貼補, 補助, 對象, 為, 何, , ] \n", - "5 [如何, 認定, 是否, 符合, 父母, 未, 就業, 家庭, 育兒, 津貼, 中, 的, ... \n", - "6 [未, 就業, 家庭, 育兒, 津貼, 條件, 中, , , 若, 一方, 無, 工作, 但... \n", - "7 [就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 所稱, 的, , ,... \n", - "8 [我, 未來, 想, 當, 合格, 托, 育人, 人員, , , 但, 又, 不是, 相關,... \n", - "9 [我, 該, 如何, 成, 為, 居家, 托, 育, 服務, 服務中心, 中心, , , 原... \n", - "10 [親戚, 親戚朋友, 朋友, 托, 育, , , 是否, 也, 可以, 領, 補助, 呢, , ] \n", - "11 [就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 申請, 應, 備, ... \n", - "12 [育兒, 津貼, 貼補, 補助, 內容, 為, 為, 何, , ] \n", - "13 [誰, 可以, 提出, 育兒, 津貼, 申請, 手續, , ] \n", - "14 [托, 育, 費用, 補助, 標準, 為, 何, , ] \n", - "15 [托, 育, 費用, 補助, 對象, 為, 何, , ] \n", - "16 [育兒, 津貼, 貼補, 補助, 補助金, 金額, 多少, , ] \n", - "17 [請, 提供, 本年, 本年度, 年度, 保, 母系, 系統, , , 居家, 托, 育, ... \n", - "18 [育兒, 津貼, 受理, 單位, 為, 何, , , 審, 核, 程序, 為, 何, , ] \n", - "19 [要, 申請, 托, 育, 補助, 一定, 定要, 找, 加入, 居家, 托, 育, 服務,... \n", - "20 [就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 申請, 對象, 及,... \n", - "21 [就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 審, 核, 程序, ... \n", - "22 [托, 育, 補助, 補助金, 金額, 多少, , , 補助, 期間, 多長, , ] \n", - "23 [免費, 育兒, 指導, 的, 內容, 是, 什麼, , ] \n", - "24 [育兒, 指導, 指導員, 到, 府, 可, 提供, 哪些, 服務, 內容, , , ] \n", - "25 [目前, 有, 領取, 育嬰, 留職, 留職停薪, 停薪, 不能, 請領, 育兒, 津貼, ... \n", - "26 [居家, 式, 托, 育人, 人員, , , 原, 保, 母, , , 或, 親屬, 保, ... \n", - "27 [申請, 育兒, 指導, 服務, 務要, 符合, 怎樣, 的, 資格, , ] \n", - "28 [如何, 申請, 育兒, 指導, 服務, , ] \n", - "29 [育兒, 指導, 服務, 可以, 提供, 哪些, 協助, , , 全部, 免費, 服務, 嗎... \n", - ".. ... \n", - "297 [取用, 食物, 的, 時候, 要, 注意, 什麼, 嗎, , ] \n", - "298 [取用, 食物, 有限, 限制, 數量, 及, 區域, 嗎, , ] \n", - "299 [放在, 冰箱, 中, 的, 食物, 安全, 嗎, , ] \n", - "300 [續, 食, 是, 什麼, , ] \n", - "301 [什麼, 是, 愛, 享, 冰箱, , ] \n", - "302 [推動, 社區, 愛, 享, 冰箱, 的, 目的, , ] \n", - "303 [請問, 如何, 進行, 合作, 合作社, 之, 清算, , ] \n", - "304 [如何, 申請, 本市, 社會, 社會福利, 福利, 慈善, 基金, 基金會, , ] \n", - "305 [本市, 人民, 人民團體, 民團, 團體, 申請, 設立, 程序, 及, 法令, 依據, , ] \n", - "306 [請問, 有關, 關合, 合作, 合作社, 變更, 登記, 應, 備, 文件, 為, 何, , ] \n", - "307 [本市, 人民, 人民團體, 民團, 團體, 選舉, 相關, 重要, 法令, , ] \n", - "308 [請問, 合作, 合作社, 場, 解散, 之, 程序, 為, 何, , ] \n", - "309 [請問, 如何, 申請, 籌組, 組合, 合作, 合作社, 場, , ] \n", - "310 [申請, 籌組, 組人, 人民, 人民團體, 民團, 團體, 之, 條件, 為, 何, , ] \n", - "311 [何謂, 合作, 合作社, , ] \n", - "312 [如何, 申請, 設立, 財團, 財團法人, 法人, 社會, 社會福利, 福利, 慈善, 慈... \n", - "313 [申請, 申請人, 人民, 人民團體, 民團, 團體, 應, 附, 書, 表, 及, 證件,... \n", - "314 [如何, 加入, 志, 工行, 行列, , ] \n", - "315 [如何, 申請, 志, 工, 隊] \n", - "316 [如何, 申請, 志願, 服務, 紀錄, 冊] \n", - "317 [如何, 申請, 志願, 服務, 榮譽, 卡] \n", - "318 [加入, 國民, 年金, 保險, , , 經審查, 審查, 符合, 所得, 未, 達, 一定... \n", - "319 [申請, 國民, 年金, 所得, 未, 達, 一定, 定標, 標準, 資格, 認定, , ,... \n", - "320 [國民, 年金, 開辦, 時, 年滿, 65, 歲, 之一, 一般, 國民, , , 請領,... \n", - "321 [國民, 年金, 保險, 被保險人, 保險, 保險人, 如果, 是, 家庭, 家庭收入, 收... \n", - "322 [國民, 年金, 之, 實施, 對象, 為, 何, , ] \n", - "323 [國民, 年金, 金生, 生育, 給付, 從, 什麼, 時候, 開辦, , ] \n", - "324 [國民, 年金, 金生, 生育, 給付, 從, 何時, 開始, 調, 漲, , ] \n", - "325 [國民, 年金, 金生, 生育, 給付, 的, 申請, 資格, , ] \n", - "326 [國民, 年金, 所得, 未, 達, 一定, 定標, 標準, 準保, 保費, 減免, 的, ... \n", - "\n", - "[326 rows x 3 columns]" - ] - }, - "execution_count": 3, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "#前處理\n", - "all_terms = []\n", - "def preprocess(item): ##定義前處理的function\n", - " terms = [t for t in jieba.cut(item, cut_all=True)] ## 把全切分模式打開，可以比對的詞彙比較多\n", - " all_terms.extend(terms) ## 收集所有出現過的字\n", - " return terms\n", - "df_question['processed'] = df_question['question'].apply(preprocess)\n", - "df_question" - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "metadata": { - "scrolled": true - }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "['', '館內', '時會', '管道', '科系', '如有', '六個', '機構', '時間', '員', '您好', '指導員', '單親', '停薪', '諮詢服務', '使用者', '療', '服務型', '誰', '冰箱', '勞', '認定', '情形', '車輛', '車車', '生育', '救援', '費', '能力', '哪些', '設', '實施', '子女教育', '早期', '防治', '祖父母', '登記', '畫', '水災', '隊', '外', '庭子', '國內', '上課時', '目前', '合法', '扣除', '生病', '重要', '小組', '自己', '保留地', '接受', '健康', '扶助', '幫助', '加收', '卡', '經濟', '樣', '包括', '指導', '自', '手續', '育人', '眼鏡', '併', '和', '審查', '專用', '樓', '害人', '急難', '留職', '503504', '保證金', '列入', '親職', '人員', '金額', '市民', '啟', '弱勢', '養', '中心', '教室', '使用', '受理', '基本', '核發', '小孩', '大津', '休閒', '保費', '婚後', '計', '擔任', '包含', '何在', '達', '但有', '無法', '申領', '事故', '現象', '檢附', '障礙者', '器具', '請領', '算是', '朋友', '遭逢', '監護', '社會', '親戚', '式', '嗎', '遷入', '解散', '原', '托', '之前', '輔助', '所有', '均', '防治法', '區域', '所得', '耗材', '人生', '志願', '6628', '於', '手語', '一審', '提供', '急救', '位置', '出生', '保', '事業', '醫院', '親戚朋友', '計算', '現金', '生活費', '部分', '樂', '物資', '證', '贈物', '重新', '往返', '判斷', '加國', '家庭暴力', '法令', '探視', '津貼', '訓練課', '不同', '所', '志', '乘', '幼稚園', '何計', '欲', '方法', '聯絡', '組合', '人定', '過長', '當事人', '食', '財產', '再', '資格', '請安', '請假', '違反', '生火', '課程', '符合', '二手', '一般', '易成', '警察', '暴力事件', '何謂', '禮金', '日間', '法律', '合格', '關於', '但', '長青', '做', '0', '異動', '問人', '人民團體', '額度', '場所', '案件', '務要', '之', '津貼費', '巴士', '歸', '性', '個人所有', '若', '換補', '在職', '爭議', '詢問', '前', '覺得', '兒童', '經銷', '事項', '狀況', '兩', '報名', '巿', '大傷', '似乎', '間或', '完小', '何', '通常', '公車', '備案', '執照', '男生', '遭遇', '領取', '老人', '失', '老花眼', '入住', '議事', '侵害', '提出申請', '財團', '評估', '保險', '節敬', '對象', '飽', '其他', '撫養', '輔', '6', '疏忽', '要到', '基金會', '車票', '該', '收出', '戶籍', '處以', '民團', '搭載', '訓練', '場', '地址', '市區', '報告書', '籌組', '容量', '返鄉', '公共', '在職訓練', '設立', '巡迴', '食堂', '三款', '住民', '3628', '國外', '好像', '那', '龐大', '無力', '成安', '受到', '求', '保障', '需要', '的', '個', '容易', '續', '留職停薪', '社區', '老', '老年', '心服', '由', '設置', '住院', '設施', '減免', '備', '日托', '多少', '一方', '97', '之後', '長期', '不好', '4', '要領', '館', '遲緩', '加長', '證物', '公益', '500', '準備', '保留', '無', '是從', '透過', '以上', '該項', '手冊', '父母', '定義', '想要', '單親家庭', '災害', '仍', '成員', '遭到', '小時', '指標', '較', '或是', '身分', '就業者', '可否', '活動中心', '救助金', '地震', '合作社', '前夫', '什麼樣', '及其', '如何', '有生', '要以', '幾歲', '時', '遷徙', '到', '新制', '媽媽', '爸爸', '共餐', '幾個', '育', '社團', '2', '在', '了', '耳', '販運', '條件', '一里', '風險', '一名', '家庭成員', '造成', '入戶', '家庭收入', '他', '身份證', '從', '家', '以下', '業務', '為', '健', '行動', '她', '報案', '颱風', '府', '同居', '換證', '就業', '暴力', '母教', '而', '歲', '課程內容', '台', '多長', '實際', '之一', '眾', '月', '清寒', '安全', '灣', '向', '生活', '呷', '孤苦無依', '保密', '補助費', '從事', '換成', '我們', '人民', '有', '接下', '並', '人工', '發', '依', '付費', '項目', '關懷', '育兒', '人', '中低收入', '將', '工行', '居住', '何處', '申', '補助金', '檢查', '課時', '前車', '服務項目', '形如', '訴訟', '問答', '人口數', '養人', '營運', '學苑', '叫做', '小朋友', '中低', '能', '附', '目的', '置換', '互毆', '會館', '1', '何種', '開放', '進行', '找', '有限', '老花', '低', '親媽', '表格', '慈善', '心理', '礙手', '本年', '四大', '日期', '女兒', '離職', '常會', '各', '入', '更新', '期限', '親子', '髖關節', '興建', '領', '補發', '活動', '是', '發現', '問答集', '救助', '開辦', '車資', '定要', '相關', '母', '一戶', '不', '審', '公費', '給付', '申辦', '依據', '新手', '基本工資', '適合', '須', '需', '做到', '查詢', '才', '途徑', '系統', '入社', '看護', '鑑定', '例假日', '甚麼', '及', '用者', '腎', '聲', '慈善事業', '土地價格', '被保險人', '問題', '親家', '本市', '工資', '報人', '獲得', '彩券', '捐款', '友好', '快要', '惟', '借用', '教育課', '人文', '永久', '輔導', '被害人', '至多', '何不', '榮譽', '漲', '方式', '處理', '保管', '訪視', '件', '本年度', '特徵', '要付', '人事', '箱內', '年', '加害', '已', '因', '托兒', '上有', '幫忙', '可不', '上班', '強制', '據點', '外縣', '暫行', '復健', '位', '報高', '關節', '有些', '反應', '數量', '問到', '證明', '宅', '屬於', '翻譯', '懲罰', '狀況不佳', '有期', '失效', '請入', '被', '被害', '有失', '每年', '怎麼', '來', '清算', '一些', '體驗', '班上', '當事', '固定', '準保', '火災', '事情', '立案', '專車', '團體', '服務', '怎樣', '孤苦', '訂', '哪個', '被判', '捐', '祖父', '發生', '致', '文康', '補助', '生', '措施', '假牙', '那些', '遊', '大陸', '新竹市', '跟', '費用', '常聽到', '如', '加課', '窗口', '怎麼辦', '連線', '基金', '臨時', '未來', '新', '參加國', '障礙', '受害', '偶或', '變更', '預約', '老師', '年度', '滯留', '成', '諮商', '房屋', '失業', '期間', '等', '遭受', '未', '內容', '法律諮詢', '公共場所', '出嫁', '幾位', '租金', '範圍', '發給', '寄養', '安', '種類', '應計', '就', '食物', '治法', '未滿', '什麼', '補助額', '重陽', '防老', '持有', '原來', '獨居', '老花眼鏡', '減免額', '金領', '參與', '收入', '家庭', '65', '家中', '衛生', '康巴', '家園', '育嬰', '幸福', '腿', '都', '全部', '保險人', '領有', '刮', '經銷商', '人口', '訂車', '上關', '養家', '停車', '取用', '薪資', '定期', '業者', '婦女', '低收入', '一級', '社區活動', '死亡', '疑似', '一定', '應向', '扶養', '者', '醫療費', '會', '紀錄', '處理方式', '年滿', '辦理', '破舊', '暫行條例', '如家', '附近', '每月', '捐贈', '走失', '才能', '到期', '暫時', '非上', '應於', '始發', '絛', '服務業', '不可', '齒', '時候', '合請', '條例', '內', '媒合', '所稱', '居家', '中', '還要', '洽詢', '通報', '微型', '可循', '市內', '女探', '貼補', '標準', '幼童', '復', '拐杖', '重大', '公告', '英文', '保證', '次數', '執業', '還', '應該', '難事', '呢', '教育', '原因', '有關', '上課', '學習', '用到', '愛', '國民', '殘障', '淹水', '孩子', '和解', '流程', '加害人', '文件', '協助', '假日', '下來', '書', '收養', '洗', '服務中心', '陷阱', '同一', '可獲', '環境', '果實', '定標', '後', '會員', '個人', '少年', '可向', '利生', '參加', '發放', '選舉', '知本', '重陽節', '師', '監護權', '高風險', '大約', '又', '公共設施', '開始', '經審查', '得知', '搭乘', '室', '勞保', '暨', '聽到', '照顧', '行列', '輛數', '令', '事件', '有人', '親家庭', '協會', '安置', '權益', '核算', '餐食', '健康檢查', '縣市', '剛', '判定', '外籍', '行政', '預防', '唯一', '資訊', '具體', '關合', '換', '是不是', '住宿', '保護', '不能', '六個月', '法人', '證件', '我家', '日', '合身', '只能', '請問', '馬上', '市府', '聽說', '因故', '哪邊', '如何是好', '遊民', '求助', '應', '一樣', '重複', '新竹', '傷病', '同意', '沒有', '其', '至', '社會福利', '花眼', '地', '保險費', '同居人', '提出', '家境', '必須', '哪', '或', '雙親', '失手', '身份', '外縣市', '顧及', '外出', '籍', '收費', '收養人', '裡', '諮詢', '年金', '規定', '社會工作', '民', '受損', '投保', '女生', '幼稚', '最近', '當', '免費', '母系', '緊急', '丈夫', '運行', '一', '子女', '申請人', '政府', '夫妻', '土地', '各項', '組人', '核', '民家', '對於', '不符', '型態', '上班時間', '可不可以', '分', '單位', '環境衛生', '電子', '發展', '更換', '每人', '內設', '認養', '仍未', '熊', '孫子', '冊', '力', '遺失', '價格', '可能', '地價', '特殊', '如果', '敬老', '享', '配偶', '要', '具', '生活費用', '身心', '限制', '推動', '想', '我', '福利', '接下來', '何時', '10', '雙方', '手鍊', '不一', '困難', '家境清寒', '地點', '離婚', '3', '財團法人', '元', '親屬', '也', '放在', '工', '需求', '認領', '如此', '仍要', '在家', '可', '總收入', '之人', '是否', '合作', '請', '表', '前妻', '虐待', '注意', '工作人員', '報告', '工作', '報', '醫療', '金生', '園', '受害人', '申請', '作人', '不是', '治療', '可以', '程序', '出', '善事', '例假', '公會', '它', '資料', '多久', '調', '已經', '短期', '境遇', '加入', '與']\n" - ] - } - ], - "source": [ - "# 建立termindex\n", - "termindex = list(set(all_terms))\n", - "print(termindex)" - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "1012\n" - ] - } - ], - "source": [ - "# 建立IDF vector\n", - "Doc_Length = len(df_question) ## 計算出共有幾篇文章\n", - "Idf_vector = [] ## 初始化IDF向量\n", - "for term in termindex: ## 對index中的詞彙跑回圈\n", - " num_of_doc_contains_term = 0 ## 計算有機篇文章出現過這個詞彙\n", - " for terms in df_question['processed']:\n", - " if term in terms:\n", - " num_of_doc_contains_term += 1\n", - " idf = np.log(Doc_Length/num_of_doc_contains_term) ## 計算該詞彙的IDF值\n", - " Idf_vector.append(idf)\n", - "print(len(Idf_vector))" - ] - }, - { - "cell_type": "code", - "execution_count": 6, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
questionansprocessedvector
0小孩出生後應於何時申請育兒津貼?1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在...[小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
1小孩出生後應於何時申請育兒津貼?隨時提出;津貼經審查通過後，追溯自受理申請之當月起發給。兒童出生後六十日內向戶政事務所完成出...[小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
2育兒津貼申請應備文件為何?申請資料應備齊:(一)兒童之戶口名簿影本。(二)申請人之郵局存摺封面影本。(三)父母雙方身分...[育兒, 津貼, 申請, 應, 備, 文件, 為, 何, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
3若民眾夫妻雙方均失業，是否可申請家庭育兒津貼費用補助一、育兒津貼補助對象：1.育有二足歲以下兒童。2.兒童之父母至少一方因育兒需要，致未能就業者...[若, 民, 眾, 夫妻, 雙方, 均, 失業, , , 是否, 可, 申請, 家庭, 育兒...[0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
4育兒津貼補助對象為何？育兒津貼補助對象，應符合下列規定：(一)育有二足歲以下兒童。(二)兒童之父母(或監護人)至少...[育兒, 津貼, 貼補, 補助, 對象, 為, 何, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
5如何認定是否符合父母未就業家庭育兒津貼中的未就業？1.未參加勞工「就業」保險（或軍保、公保）。2.最近一年度之稅捐稽徵機關核定之薪資所得及執行...[如何, 認定, 是否, 符合, 父母, 未, 就業, 家庭, 育兒, 津貼, 中, 的, ...[0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
6未就業家庭育兒津貼條件中，若一方無工作但有投保公會勞保，請問是否可申請該項津貼，請提供該項業...針對勞工保險投保職業工會者（或雇主）等對象，一般均未參加「勞工就業保險」（自行加保者除外），...[未, 就業, 家庭, 育兒, 津貼, 條件, 中, , , 若, 一方, 無, 工作, 但...[0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...
7就業者家庭部分托育費用補助所稱的「就業者」為何？(1)受僱於政府、學校或公民營事業單位者。(2)勞動基準法所稱受雇主僱用從事工作者。(3)依...[就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 所稱, 的, , ,...[0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...
8我未來想當合格托育人員，但又不是相關科系，新竹市哪裡會開辦托育人員訓練課程呢？勞動部勞動力發展署及本府委託社團法人新竹市社區婦女關懷協會、社團法人新竹市嬰幼兒保育協會、新...[我, 未來, 想, 當, 合格, 托, 育人, 人員, , , 但, 又, 不是, 相關,...[0.28262681512484084, 0.0, 0.0, 0.0, 5.7868973...
9我該如何成為居家托育服務中心(原社區保母系統)的居家式托育人員呢？欲申請加入居家托育服務中心內的居家托育人員依規需年滿20歲並符合以下資格之一：1.取得保母技...[我, 該, 如何, 成, 為, 居家, 托, 育, 服務, 服務中心, 中心, , , 原...[0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...
10親戚朋友托育，是否也可以領補助呢？一般民眾(含寶寶的爺爺奶奶、阿姨、叔叔、舅舅等三親等以內親屬)必須上課126小時托育人員訓練...[親戚, 親戚朋友, 朋友, 托, 育, , , 是否, 也, 可以, 領, 補助, 呢, , ][0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,...
11就業者家庭部分托育費用補助申請應備文件為何？向誰申請1.補助申請表。2.最近戶籍謄本或戶口名簿影本。3.郵局存簿封面影本。4.托育契約書（親屬托...[就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 申請, 應, 備, ...[0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
12育兒津貼補助內容為為何？本津貼補助對象，應符合下列規定：(一)育有2足歲以下兒童。(二)兒童之父母(或監護人)至少一...[育兒, 津貼, 貼補, 補助, 內容, 為, 為, 何, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
13誰可以提出育兒津貼申請手續？(一)申請人資格規定如下：1.育有二足歲以下兒童。2.兒童之父母(或監護人)至少一方因育兒需...[誰, 可以, 提出, 育兒, 津貼, 申請, 手續, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
14托育費用補助標準為何？補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托...[托, 育, 費用, 補助, 標準, 為, 何, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
15托育費用補助對象為何？補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總...[托, 育, 費用, 補助, 對象, 為, 何, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
16育兒津貼補助金額多少？育兒津貼補助金額依家庭狀況而定，補助金額如下：(一)低收入戶：每名兒童每月補助新臺幣5000...[育兒, 津貼, 貼補, 補助, 補助金, 金額, 多少, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
17請提供本年度保母系統(居家托育服務中心)相關資訊為何?目前本市居家托育服務中心承辦單位及聯絡電話如下：東區居家托育服務中心承辦單位：新竹市私立光復...[請, 提供, 本年, 本年度, 年度, 保, 母系, 系統, , , 居家, 托, 育, ...[0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...
18育兒津貼受理單位為何？審核程序為何？本津貼申領及發放程序規定如下：(一)由申請人檢具相關證明文件向兒童戶籍地之區公所提出申請。(...[育兒, 津貼, 受理, 單位, 為, 何, , , 審, 核, 程序, 為, 何, , ][0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,...
19要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服...[要, 申請, 托, 育, 補助, 一定, 定要, 找, 加入, 居家, 托, 育, 服務,...[0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
20就業者家庭部分托育費用補助申請對象及審核程序為何？就業者家庭部分托育費用補助的實施對象為：1.父母(或監護人)雙方或單親一方皆就業，或父母一方...[就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 申請, 對象, 及,...[0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
21就業者家庭部分托育費用補助審核程序為何？父母需於托育事實發生日起15日內，檢齊應備文件送交幼兒托育地點之社區保母系統或托嬰中心，其補...[就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 審, 核, 程序, ...[0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
22托育補助金額多少，補助期間多長？依各類家庭條件及送托托育人員資格，其托育補助金額說明如下：(一)將嬰幼兒送至托育服務中心中有...[托, 育, 補助, 補助金, 金額, 多少, , , 補助, 期間, 多長, , ][0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,...
23免費育兒指導的內容是什麼？政府建構友善的托育環境，除了提供托育費用及育兒津貼的經濟補助外，仍然重視親職教育，所以提供免...[免費, 育兒, 指導, 的, 內容, 是, 什麼, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
24育兒指導員到府可提供哪些服務內容？（一）提供嬰幼兒進食、遊戲、睡眠及幼兒活動路線等適性與安全諮詢。（二）提供嬰幼兒穿衣、換尿布...[育兒, 指導, 指導員, 到, 府, 可, 提供, 哪些, 服務, 內容, , , ][0.14131340756242042, 0.0, 0.0, 0.0, 0.0, 0.0,...
25目前有領取育嬰留職停薪不能請領育兒津貼，但育嬰留職停薪只能請領六個月，之後孩子仍未滿2歲，還...領取育嬰留職停薪津貼的六個月內，同一孩童不得再領取育兒津貼。但如果在育嬰留職停薪期間但未領取...[目前, 有, 領取, 育嬰, 留職, 留職停薪, 停薪, 不能, 請領, 育兒, 津貼, ...[0.37683575349978776, 0.0, 0.0, 0.0, 0.0, 0.0,...
26居家式托育人員(原保母)或親屬保母每人至多可以照顧幾位幼童？居家式托育人員或親屬保母每人至多照顧兒童(含托育人員本人之幼兒)4人，其中未滿2歲者最多2人...[居家, 式, 托, 育人, 人員, , , 原, 保, 母, , , 或, 親屬, 保, ...[0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...
27申請育兒指導服務要符合怎樣的資格？只要符合下列資格之一，即可申請：（一）設籍且實際居住本市，家有0-2歲以下幼兒並有育兒指導需...[申請, 育兒, 指導, 服務, 務要, 符合, 怎樣, 的, 資格, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
28如何申請育兒指導服務？（一）、電話諮詢且評估到府指導需求，服務電話：03-5396000。（二）、新竹市政府社會處...[如何, 申請, 育兒, 指導, 服務, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
29育兒指導服務可以提供哪些協助？全部免費服務嗎？育兒指導服務是新竹市政府社會處主辦的免費服務方案；提供電話親子教養諮詢，且評估實際需求後，由...[育兒, 指導, 服務, 可以, 提供, 哪些, 協助, , , 全部, 免費, 服務, 嗎...[0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,...
...............
297取用食物的時候要注意什麼嗎？106-12-31[取用, 食物, 的, 時候, 要, 注意, 什麼, 嗎, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
298取用食物有限制數量及區域嗎？106-12-31[取用, 食物, 有限, 限制, 數量, 及, 區域, 嗎, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
299放在冰箱中的食物安全嗎?106-12-31[放在, 冰箱, 中, 的, 食物, 安全, 嗎, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
300續食是什麼?106-12-31[續, 食, 是, 什麼, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
301什麼是愛享冰箱?106-12-31[什麼, 是, 愛, 享, 冰箱, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
302推動社區愛享冰箱的目的?106-12-31[推動, 社區, 愛, 享, 冰箱, 的, 目的, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
303請問如何進行合作社之清算？合作社（場）於奉令解散或陳准解散登記後，應即進行清算手續。合作社（場）之清算登記，旨在使得合...[請問, 如何, 進行, 合作, 合作社, 之, 清算, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
304如何申請本市社會福利慈善基金會？業務範圍僅在新竹市者，向新竹市政府社會處申請，其設立基金最低數額為一千萬元以上。應備文件（各...[如何, 申請, 本市, 社會, 社會福利, 福利, 慈善, 基金, 基金會, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
305本市人民團體申請設立程序及法令依據？依據人民團體法第8條規定人民團體之組織，應由發起人檢具申請書、章程草案及發起人名冊，向主管機...[本市, 人民, 人民團體, 民團, 團體, 申請, 設立, 程序, 及, 法令, 依據, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
306請問有關合作社變更登記應備文件為何？1.變更登記申請書。2.入社社員名冊。3.退（出）社社員名冊。4.年度社員大會紀錄(含改選理...[請問, 有關, 關合, 合作, 合作社, 變更, 登記, 應, 備, 文件, 為, 何, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
307本市人民團體選舉相關重要法令？人民團體選舉罷免辦法第5條人民團體理事、監事及會員代表之選舉或罷免，應由理事會在召開會議十五...[本市, 人民, 人民團體, 民團, 團體, 選舉, 相關, 重要, 法令, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
308請問合作社場解散之程序為何？合作社因左列各款情事之一而解散：1.章程所定解散之事由發生。2.社員大會之解散決議（應有全體...[請問, 合作, 合作社, 場, 解散, 之, 程序, 為, 何, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
309請問如何申請籌組合作社場？有共同需要實行合作的人，如要組織合作社（場）時，應有七人以上發起人，發起前應先洽詢主管機關索...[請問, 如何, 申請, 籌組, 組合, 合作, 合作社, 場, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
310申請籌組人民團體之條件為何？一、申請條件：須設籍於本市之市民30人以上發起二、發起人之基本資格：須本市市民、年滿二十歲，...[申請, 籌組, 組人, 人民, 人民團體, 民團, 團體, 之, 條件, 為, 何, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
311何謂合作社？所謂合作社係指依平等原則，在互助組織之基礎上，以共同經營方法，謀社員經濟之利益與生活之改善，...[何謂, 合作, 合作社, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
312如何申請設立財團法人社會福利慈善事業基金會？一、應檢附下列文件：1.申請書(一份)2.基金會概況表(一份)3.基金會籌備會議記錄(四份)...[如何, 申請, 設立, 財團, 財團法人, 法人, 社會, 社會福利, 福利, 慈善, 慈...[0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
313申請人民團體應附書表及證件為何？一、社會團體：1、新竹市社會團體申請書2、新竹市社會團體章程3、新竹市社會團體發起人名冊4、...[申請, 申請人, 人民, 人民團體, 民團, 團體, 應, 附, 書, 表, 及, 證件,...[0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
314如何加入志工行列？參考本府所編製之志願服務資源簡章，瞭解各類別志工隊服務內容及所需志工資格，再依本身志趣，逕向...[如何, 加入, 志, 工行, 行列, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
315如何申請志工隊依志願服務法所訂志願服務運用單位係指運用志工之機關、機構、學校、法人或經政府立案之團體。符合...[如何, 申請, 志, 工, 隊][0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...
316如何申請志願服務紀錄冊需先完成12小時之志願服務基礎訓練和特殊訓練。完成上述訓練者，由志願服務運用單位志願服務運用...[如何, 申請, 志願, 服務, 紀錄, 冊][0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...
317如何申請志願服務榮譽卡志工服務年資滿3年、服務時數達300小時以上者，得檢具1吋半身照2張、服務紀錄冊影本、申請表...[如何, 申請, 志願, 服務, 榮譽, 卡][0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...
318加入國民年金保險，經審查符合所得未達一定標準者，如有戶籍遷徙，是否應重新申請？國民年金保費減免係依被保險人當月底戶籍所在地直轄市、縣（市）主管機關為準。故申請符合所得未達...[加入, 國民, 年金, 保險, , , 經審查, 審查, 符合, 所得, 未, 達, 一定...[0.37683575349978776, 0.0, 0.0, 0.0, 0.0, 4.68...
319申請國民年金所得未達一定標準資格認定，應向哪一單位提出申請？請攜帶申請人身分證、印章及全戶戶口名簿等資料至戶籍所在地區公所社政課提出申請。[申請, 國民, 年金, 所得, 未, 達, 一定, 定標, 標準, 資格, 認定, , ,...[0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,...
320國民年金開辦時年滿65歲之一般國民，請領老年基本保證年金之資格為何？國民年金開辦時年滿65歲國民，在國內設有戶籍，且於最近3年內每年居住超過183日，而無下列情...[國民, 年金, 開辦, 時, 年滿, 65, 歲, 之一, 一般, 國民, , , 請領,...[0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,...
321國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於...[國民, 年金, 保險, 被保險人, 保險, 保險人, 如果, 是, 家庭, 家庭收入, 收...[0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,...
322國民年金之實施對象為何？1.年滿25歲、未滿65歲，在國內設有戶籍，且未參加勞保、農保、公教保、軍保之國民。2.國民...[國民, 年金, 之, 實施, 對象, 為, 何, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
323國民年金生育給付從什麼時候開辦？自100年7月1日實施。[國民, 年金, 金生, 生育, 給付, 從, 什麼, 時候, 開辦, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
324國民年金生育給付從何時開始調漲？自104年12月18日起調漲，由1個月的國民年金月投保金額調整至2個月。105年月投保金額為...[國民, 年金, 金生, 生育, 給付, 從, 何時, 開始, 調, 漲, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
325國民年金生育給付的申請資格？1.生產期間需有國民年金保險身分。2.自100年7月1日後生產。3.生產日起算5年內需向各地...[國民, 年金, 金生, 生育, 給付, 的, 申請, 資格, , ][0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
326國民年金所得未達一定標準保費減免的申請方式？1.向戶籍所在地區公所申請及洽詢。2.攜帶身分證、印章、戶口名簿。3.全家人口之退休俸、學生...[國民, 年金, 所得, 未, 達, 一定, 定標, 標準, 準保, 保費, 減免, 的, ...[0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,...
\n", - "

326 rows × 4 columns

\n", - "
" - ], - "text/plain": [ - " question \\\n", - "0 小孩出生後應於何時申請育兒津貼? \n", - "1 小孩出生後應於何時申請育兒津貼? \n", - "2 育兒津貼申請應備文件為何? \n", - "3 若民眾夫妻雙方均失業，是否可申請家庭育兒津貼費用補助 \n", - "4 育兒津貼補助對象為何？ \n", - "5 如何認定是否符合父母未就業家庭育兒津貼中的未就業？ \n", - "6 未就業家庭育兒津貼條件中，若一方無工作但有投保公會勞保，請問是否可申請該項津貼，請提供該項業... \n", - "7 就業者家庭部分托育費用補助所稱的「就業者」為何？ \n", - "8 我未來想當合格托育人員，但又不是相關科系，新竹市哪裡會開辦托育人員訓練課程呢？ \n", - "9 我該如何成為居家托育服務中心(原社區保母系統)的居家式托育人員呢？ \n", - "10 親戚朋友托育，是否也可以領補助呢？ \n", - "11 就業者家庭部分托育費用補助申請應備文件為何？向誰申請 \n", - "12 育兒津貼補助內容為為何？ \n", - "13 誰可以提出育兒津貼申請手續？ \n", - "14 托育費用補助標準為何？ \n", - "15 托育費用補助對象為何？ \n", - "16 育兒津貼補助金額多少？ \n", - "17 請提供本年度保母系統(居家托育服務中心)相關資訊為何? \n", - "18 育兒津貼受理單位為何？審核程序為何？ \n", - "19 要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？ \n", - "20 就業者家庭部分托育費用補助申請對象及審核程序為何？ \n", - "21 就業者家庭部分托育費用補助審核程序為何？ \n", - "22 托育補助金額多少，補助期間多長？ \n", - "23 免費育兒指導的內容是什麼？ \n", - "24 育兒指導員到府可提供哪些服務內容？ \n", - "25 目前有領取育嬰留職停薪不能請領育兒津貼，但育嬰留職停薪只能請領六個月，之後孩子仍未滿2歲，還... \n", - "26 居家式托育人員(原保母)或親屬保母每人至多可以照顧幾位幼童？ \n", - "27 申請育兒指導服務要符合怎樣的資格？ \n", - "28 如何申請育兒指導服務？ \n", - "29 育兒指導服務可以提供哪些協助？全部免費服務嗎？ \n", - ".. ... \n", - "297 取用食物的時候要注意什麼嗎？ \n", - "298 取用食物有限制數量及區域嗎？ \n", - "299 放在冰箱中的食物安全嗎? \n", - "300 續食是什麼? \n", - "301 什麼是愛享冰箱? \n", - "302 推動社區愛享冰箱的目的? \n", - "303 請問如何進行合作社之清算？ \n", - "304 如何申請本市社會福利慈善基金會？ \n", - "305 本市人民團體申請設立程序及法令依據？ \n", - "306 請問有關合作社變更登記應備文件為何？ \n", - "307 本市人民團體選舉相關重要法令？ \n", - "308 請問合作社場解散之程序為何？ \n", - "309 請問如何申請籌組合作社場？ \n", - "310 申請籌組人民團體之條件為何？ \n", - "311 何謂合作社？ \n", - "312 如何申請設立財團法人社會福利慈善事業基金會？ \n", - "313 申請人民團體應附書表及證件為何？ \n", - "314 如何加入志工行列？ \n", - "315 如何申請志工隊 \n", - "316 如何申請志願服務紀錄冊 \n", - "317 如何申請志願服務榮譽卡 \n", - "318 加入國民年金保險，經審查符合所得未達一定標準者，如有戶籍遷徙，是否應重新申請？ \n", - "319 申請國民年金所得未達一定標準資格認定，應向哪一單位提出申請？ \n", - "320 國民年金開辦時年滿65歲之一般國民，請領老年基本保證年金之資格為何？ \n", - "321 國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？ \n", - "322 國民年金之實施對象為何？ \n", - "323 國民年金生育給付從什麼時候開辦？ \n", - "324 國民年金生育給付從何時開始調漲？ \n", - "325 國民年金生育給付的申請資格？ \n", - "326 國民年金所得未達一定標準保費減免的申請方式？ \n", - "\n", - " ans \\\n", - "0 1.幼兒家長在戶政事務所完成新生兒出生登記後，即可向所轄區公所社政課提出育兒津貼申請。2.在... \n", - "1 隨時提出;津貼經審查通過後，追溯自受理申請之當月起發給。兒童出生後六十日內向戶政事務所完成出... \n", - "2 申請資料應備齊:(一)兒童之戶口名簿影本。(二)申請人之郵局存摺封面影本。(三)父母雙方身分... \n", - "3 一、育兒津貼補助對象：1.育有二足歲以下兒童。2.兒童之父母至少一方因育兒需要，致未能就業者... \n", - "4 育兒津貼補助對象，應符合下列規定：(一)育有二足歲以下兒童。(二)兒童之父母(或監護人)至少... \n", - "5 1.未參加勞工「就業」保險（或軍保、公保）。2.最近一年度之稅捐稽徵機關核定之薪資所得及執行... \n", - "6 針對勞工保險投保職業工會者（或雇主）等對象，一般均未參加「勞工就業保險」（自行加保者除外），... \n", - "7 (1)受僱於政府、學校或公民營事業單位者。(2)勞動基準法所稱受雇主僱用從事工作者。(3)依... \n", - "8 勞動部勞動力發展署及本府委託社團法人新竹市社區婦女關懷協會、社團法人新竹市嬰幼兒保育協會、新... \n", - "9 欲申請加入居家托育服務中心內的居家托育人員依規需年滿20歲並符合以下資格之一：1.取得保母技... \n", - "10 一般民眾(含寶寶的爺爺奶奶、阿姨、叔叔、舅舅等三親等以內親屬)必須上課126小時托育人員訓練... \n", - "11 1.補助申請表。2.最近戶籍謄本或戶口名簿影本。3.郵局存簿封面影本。4.托育契約書（親屬托... \n", - "12 本津貼補助對象，應符合下列規定：(一)育有2足歲以下兒童。(二)兒童之父母(或監護人)至少一... \n", - "13 (一)申請人資格規定如下：1.育有二足歲以下兒童。2.兒童之父母(或監護人)至少一方因育兒需... \n", - "14 補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托... \n", - "15 補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總... \n", - "16 育兒津貼補助金額依家庭狀況而定，補助金額如下：(一)低收入戶：每名兒童每月補助新臺幣5000... \n", - "17 目前本市居家托育服務中心承辦單位及聯絡電話如下：東區居家托育服務中心承辦單位：新竹市私立光復... \n", - "18 本津貼申領及發放程序規定如下：(一)由申請人檢具相關證明文件向兒童戶籍地之區公所提出申請。(... \n", - "19 依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服... \n", - "20 就業者家庭部分托育費用補助的實施對象為：1.父母(或監護人)雙方或單親一方皆就業，或父母一方... \n", - "21 父母需於托育事實發生日起15日內，檢齊應備文件送交幼兒托育地點之社區保母系統或托嬰中心，其補... \n", - "22 依各類家庭條件及送托托育人員資格，其托育補助金額說明如下：(一)將嬰幼兒送至托育服務中心中有... \n", - "23 政府建構友善的托育環境，除了提供托育費用及育兒津貼的經濟補助外，仍然重視親職教育，所以提供免... \n", - "24 （一）提供嬰幼兒進食、遊戲、睡眠及幼兒活動路線等適性與安全諮詢。（二）提供嬰幼兒穿衣、換尿布... \n", - "25 領取育嬰留職停薪津貼的六個月內，同一孩童不得再領取育兒津貼。但如果在育嬰留職停薪期間但未領取... \n", - "26 居家式托育人員或親屬保母每人至多照顧兒童(含托育人員本人之幼兒)4人，其中未滿2歲者最多2人... \n", - "27 只要符合下列資格之一，即可申請：（一）設籍且實際居住本市，家有0-2歲以下幼兒並有育兒指導需... \n", - "28 （一）、電話諮詢且評估到府指導需求，服務電話：03-5396000。（二）、新竹市政府社會處... \n", - "29 育兒指導服務是新竹市政府社會處主辦的免費服務方案；提供電話親子教養諮詢，且評估實際需求後，由... \n", - ".. ... \n", - "297 106-12-31 \n", - "298 106-12-31 \n", - "299 106-12-31 \n", - "300 106-12-31 \n", - "301 106-12-31 \n", - "302 106-12-31 \n", - "303 合作社（場）於奉令解散或陳准解散登記後，應即進行清算手續。合作社（場）之清算登記，旨在使得合... \n", - "304 業務範圍僅在新竹市者，向新竹市政府社會處申請，其設立基金最低數額為一千萬元以上。應備文件（各... \n", - "305 依據人民團體法第8條規定人民團體之組織，應由發起人檢具申請書、章程草案及發起人名冊，向主管機... \n", - "306 1.變更登記申請書。2.入社社員名冊。3.退（出）社社員名冊。4.年度社員大會紀錄(含改選理... \n", - "307 人民團體選舉罷免辦法第5條人民團體理事、監事及會員代表之選舉或罷免，應由理事會在召開會議十五... \n", - "308 合作社因左列各款情事之一而解散：1.章程所定解散之事由發生。2.社員大會之解散決議（應有全體... \n", - "309 有共同需要實行合作的人，如要組織合作社（場）時，應有七人以上發起人，發起前應先洽詢主管機關索... \n", - "310 一、申請條件：須設籍於本市之市民30人以上發起二、發起人之基本資格：須本市市民、年滿二十歲，... \n", - "311 所謂合作社係指依平等原則，在互助組織之基礎上，以共同經營方法，謀社員經濟之利益與生活之改善，... \n", - "312 一、應檢附下列文件：1.申請書(一份)2.基金會概況表(一份)3.基金會籌備會議記錄(四份)... \n", - "313 一、社會團體：1、新竹市社會團體申請書2、新竹市社會團體章程3、新竹市社會團體發起人名冊4、... \n", - "314 參考本府所編製之志願服務資源簡章，瞭解各類別志工隊服務內容及所需志工資格，再依本身志趣，逕向... \n", - "315 依志願服務法所訂志願服務運用單位係指運用志工之機關、機構、學校、法人或經政府立案之團體。符合... \n", - "316 需先完成12小時之志願服務基礎訓練和特殊訓練。完成上述訓練者，由志願服務運用單位志願服務運用... \n", - "317 志工服務年資滿3年、服務時數達300小時以上者，得檢具1吋半身照2張、服務紀錄冊影本、申請表... \n", - "318 國民年金保費減免係依被保險人當月底戶籍所在地直轄市、縣（市）主管機關為準。故申請符合所得未達... \n", - "319 請攜帶申請人身分證、印章及全戶戶口名簿等資料至戶籍所在地區公所社政課提出申請。 \n", - "320 國民年金開辦時年滿65歲國民，在國內設有戶籍，且於最近3年內每年居住超過183日，而無下列情... \n", - "321 國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於... \n", - "322 1.年滿25歲、未滿65歲，在國內設有戶籍，且未參加勞保、農保、公教保、軍保之國民。2.國民... \n", - "323 自100年7月1日實施。 \n", - "324 自104年12月18日起調漲，由1個月的國民年金月投保金額調整至2個月。105年月投保金額為... \n", - "325 1.生產期間需有國民年金保險身分。2.自100年7月1日後生產。3.生產日起算5年內需向各地... \n", - "326 1.向戶籍所在地區公所申請及洽詢。2.攜帶身分證、印章、戶口名簿。3.全家人口之退休俸、學生... \n", - "\n", - " processed \\\n", - "0 [小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ] \n", - "1 [小孩, 出生, 後, 應於, 何時, 申請, 育兒, 津貼, , ] \n", - "2 [育兒, 津貼, 申請, 應, 備, 文件, 為, 何, , ] \n", - "3 [若, 民, 眾, 夫妻, 雙方, 均, 失業, , , 是否, 可, 申請, 家庭, 育兒... \n", - "4 [育兒, 津貼, 貼補, 補助, 對象, 為, 何, , ] \n", - "5 [如何, 認定, 是否, 符合, 父母, 未, 就業, 家庭, 育兒, 津貼, 中, 的, ... \n", - "6 [未, 就業, 家庭, 育兒, 津貼, 條件, 中, , , 若, 一方, 無, 工作, 但... \n", - "7 [就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 所稱, 的, , ,... \n", - "8 [我, 未來, 想, 當, 合格, 托, 育人, 人員, , , 但, 又, 不是, 相關,... \n", - "9 [我, 該, 如何, 成, 為, 居家, 托, 育, 服務, 服務中心, 中心, , , 原... \n", - "10 [親戚, 親戚朋友, 朋友, 托, 育, , , 是否, 也, 可以, 領, 補助, 呢, , ] \n", - "11 [就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 申請, 應, 備, ... \n", - "12 [育兒, 津貼, 貼補, 補助, 內容, 為, 為, 何, , ] \n", - "13 [誰, 可以, 提出, 育兒, 津貼, 申請, 手續, , ] \n", - "14 [托, 育, 費用, 補助, 標準, 為, 何, , ] \n", - "15 [托, 育, 費用, 補助, 對象, 為, 何, , ] \n", - "16 [育兒, 津貼, 貼補, 補助, 補助金, 金額, 多少, , ] \n", - "17 [請, 提供, 本年, 本年度, 年度, 保, 母系, 系統, , , 居家, 托, 育, ... \n", - "18 [育兒, 津貼, 受理, 單位, 為, 何, , , 審, 核, 程序, 為, 何, , ] \n", - "19 [要, 申請, 托, 育, 補助, 一定, 定要, 找, 加入, 居家, 托, 育, 服務,... \n", - "20 [就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 申請, 對象, 及,... \n", - "21 [就業, 就業者, 業者, 家庭, 部分, 托, 育, 費用, 補助, 審, 核, 程序, ... \n", - "22 [托, 育, 補助, 補助金, 金額, 多少, , , 補助, 期間, 多長, , ] \n", - "23 [免費, 育兒, 指導, 的, 內容, 是, 什麼, , ] \n", - "24 [育兒, 指導, 指導員, 到, 府, 可, 提供, 哪些, 服務, 內容, , , ] \n", - "25 [目前, 有, 領取, 育嬰, 留職, 留職停薪, 停薪, 不能, 請領, 育兒, 津貼, ... \n", - "26 [居家, 式, 托, 育人, 人員, , , 原, 保, 母, , , 或, 親屬, 保, ... \n", - "27 [申請, 育兒, 指導, 服務, 務要, 符合, 怎樣, 的, 資格, , ] \n", - "28 [如何, 申請, 育兒, 指導, 服務, , ] \n", - "29 [育兒, 指導, 服務, 可以, 提供, 哪些, 協助, , , 全部, 免費, 服務, 嗎... \n", - ".. ... \n", - "297 [取用, 食物, 的, 時候, 要, 注意, 什麼, 嗎, , ] \n", - "298 [取用, 食物, 有限, 限制, 數量, 及, 區域, 嗎, , ] \n", - "299 [放在, 冰箱, 中, 的, 食物, 安全, 嗎, , ] \n", - "300 [續, 食, 是, 什麼, , ] \n", - "301 [什麼, 是, 愛, 享, 冰箱, , ] \n", - "302 [推動, 社區, 愛, 享, 冰箱, 的, 目的, , ] \n", - "303 [請問, 如何, 進行, 合作, 合作社, 之, 清算, , ] \n", - "304 [如何, 申請, 本市, 社會, 社會福利, 福利, 慈善, 基金, 基金會, , ] \n", - "305 [本市, 人民, 人民團體, 民團, 團體, 申請, 設立, 程序, 及, 法令, 依據, , ] \n", - "306 [請問, 有關, 關合, 合作, 合作社, 變更, 登記, 應, 備, 文件, 為, 何, , ] \n", - "307 [本市, 人民, 人民團體, 民團, 團體, 選舉, 相關, 重要, 法令, , ] \n", - "308 [請問, 合作, 合作社, 場, 解散, 之, 程序, 為, 何, , ] \n", - "309 [請問, 如何, 申請, 籌組, 組合, 合作, 合作社, 場, , ] \n", - "310 [申請, 籌組, 組人, 人民, 人民團體, 民團, 團體, 之, 條件, 為, 何, , ] \n", - "311 [何謂, 合作, 合作社, , ] \n", - "312 [如何, 申請, 設立, 財團, 財團法人, 法人, 社會, 社會福利, 福利, 慈善, 慈... \n", - "313 [申請, 申請人, 人民, 人民團體, 民團, 團體, 應, 附, 書, 表, 及, 證件,... \n", - "314 [如何, 加入, 志, 工行, 行列, , ] \n", - "315 [如何, 申請, 志, 工, 隊] \n", - "316 [如何, 申請, 志願, 服務, 紀錄, 冊] \n", - "317 [如何, 申請, 志願, 服務, 榮譽, 卡] \n", - "318 [加入, 國民, 年金, 保險, , , 經審查, 審查, 符合, 所得, 未, 達, 一定... \n", - "319 [申請, 國民, 年金, 所得, 未, 達, 一定, 定標, 標準, 資格, 認定, , ,... \n", - "320 [國民, 年金, 開辦, 時, 年滿, 65, 歲, 之一, 一般, 國民, , , 請領,... \n", - "321 [國民, 年金, 保險, 被保險人, 保險, 保險人, 如果, 是, 家庭, 家庭收入, 收... \n", - "322 [國民, 年金, 之, 實施, 對象, 為, 何, , ] \n", - "323 [國民, 年金, 金生, 生育, 給付, 從, 什麼, 時候, 開辦, , ] \n", - "324 [國民, 年金, 金生, 生育, 給付, 從, 何時, 開始, 調, 漲, , ] \n", - "325 [國民, 年金, 金生, 生育, 給付, 的, 申請, 資格, , ] \n", - "326 [國民, 年金, 所得, 未, 達, 一定, 定標, 標準, 準保, 保費, 減免, 的, ... \n", - "\n", - " vector \n", - "0 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "1 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "2 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "3 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "4 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "5 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "6 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "7 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "8 [0.28262681512484084, 0.0, 0.0, 0.0, 5.7868973... \n", - "9 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "10 [0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "11 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "12 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "13 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "14 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "15 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "16 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "17 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "18 [0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "19 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "20 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "21 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "22 [0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "23 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "24 [0.14131340756242042, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "25 [0.37683575349978776, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "26 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "27 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "28 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "29 [0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - ".. ... \n", - "297 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "298 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "299 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "300 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "301 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "302 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "303 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "304 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "305 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "306 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "307 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "308 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "309 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "310 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "311 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "312 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "313 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "314 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "315 [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... \n", - "316 [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... \n", - "317 [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... \n", - "318 [0.37683575349978776, 0.0, 0.0, 0.0, 0.0, 4.68... \n", - "319 [0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "320 [0.18841787674989388, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "321 [0.28262681512484084, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "322 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "323 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "324 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "325 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "326 [0.09420893837494694, 0.0, 0.0, 0.0, 0.0, 0.0,... \n", - "\n", - "[326 rows x 4 columns]" - ] - }, - "execution_count": 6, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "# 建立document vector\n", - "def terms_to_vector(terms): ## 定義把terms轉換成向量的function\n", - " vector = np.zeros_like(termindex, dtype=np.float32) ## 建立一條與termsindex等長、但值全部為零的向量\n", - " for term in terms:\n", - " if term in termindex: \n", - " idx = termindex.index(term) ## 測試時如果有字沒有在索引中，需要保護\n", - " vector[idx] += 1 ## 計算term frequency\n", - " vector = vector * Idf_vector ## 如果兩個vector的型別都是np.array，把兩條vector相乘，就會自動把向量中的每一個元素成在一起，建立出一條新的向量\n", - " return vector\n", - "df_question['vector'] = df_question['processed'].apply(terms_to_vector) ## 將上面定義的function，套用在每一筆資料的terms欄位上\n", - "df_question" - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "小孩出生後應於何時申請育兒津貼?\n", - "育兒津貼申請應備文件為何?\n" - ] - }, - { - "data": { - "text/plain": [ - "0.20322784793773094" - ] - }, - "execution_count": 7, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "from numpy.linalg import norm\n", - "def cosine_similarity(vector1, vector2): ## 定義cosine相似度的計算公式\n", - " score = np.dot(vector1, vector2) / (norm(vector1) * norm(vector2))\n", - " return score\n", - "\n", - "sentence1 = df_question.loc[0] ##取出第零個的問題\n", - "sentence2 = df_question.loc[2] ##取出第二個的問題\n", - "print(sentence1['question'])\n", - "print(sentence2['question'])\n", - "cosine_similarity(sentence1['vector'], sentence2['vector']) ##計算兩者的相似度" - ] - }, - { - "cell_type": "code", - "execution_count": 12, - "metadata": { - "collapsed": true - }, - "outputs": [], - "source": [ - "def retrieve(testing_sentence, return_num=3): ## 定義出檢索引擎\n", - " testing_vector = terms_to_vector(preprocess(testing_sentence)) ## 把剛剛的前處理、轉換成向量的function，應用在使用者輸入的問題上\n", - " score_dict = {} ## 準備把每一個問題對應到使用者問題的cosine分數記錄下來\n", - " for idx, vec in enumerate(df_question['vector']): ## 計算每一個問題與使用者問題的cosine分數\n", - " score = cosine_similarity(testing_vector, vec)\n", - " score_dict[idx] = score\n", - " idxs = np.array(sorted(score_dict.items(), key=lambda x:x[1], reverse=True))[:return_num, 0] ##排序出最相關的前N個問題的row index\n", - " return df_question.loc[idxs, ['question', 'ans']]" - ] - }, - { - "cell_type": "code", - "execution_count": 14, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "請輸入您的問題?老人年金\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
questionans
100.0我已經年滿65歲領有國民年金老人年金及基本保證年金3628元，因家境清寒還可以再申請中低收入...不可以。所有的老人津貼，例如：國民年金、老農津貼、榮家院外就養金、中低收入老人生活津貼等只能...
111.0新竹市老人一般可領老人津貼6628元，該如何申請？新竹市一般老人領取國民年金基本保證年金3628元加上本市發放之安老津貼3000元，合計662...
321.0國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於...
\n", - "
" - ], - "text/plain": [ - " question \\\n", - "100.0 我已經年滿65歲領有國民年金老人年金及基本保證年金3628元，因家境清寒還可以再申請中低收入... \n", - "111.0 新竹市老人一般可領老人津貼6628元，該如何申請？ \n", - "321.0 國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？ \n", - "\n", - " ans \n", - "100.0 不可以。所有的老人津貼，例如：國民年金、老農津貼、榮家院外就養金、中低收入老人生活津貼等只能... \n", - "111.0 新竹市一般老人領取國民年金基本保證年金3628元加上本市發放之安老津貼3000元，合計662... \n", - "321.0 國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於... " - ] - }, - "execution_count": 14, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "query = input('請輸入您的問題?')\n", - "retrieve(query)" - ] - }, - { - "cell_type": "code", - "execution_count": 15, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "請輸入您的問題?托育\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
questionans
15.0托育費用補助對象為何？補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總...
14.0托育費用補助標準為何？補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托...
19.0要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服...
\n", - "
" - ], - "text/plain": [ - " question \\\n", - "15.0 托育費用補助對象為何？ \n", - "14.0 托育費用補助標準為何？ \n", - "19.0 要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？ \n", - "\n", - " ans \n", - "15.0 補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總... \n", - "14.0 補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托... \n", - "19.0 依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服... " - ] - }, - "execution_count": 15, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "query = input('請輸入您的問題?')\n", - "retrieve(query)" - ] - }, - { - "cell_type": "code", - "execution_count": 16, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "請輸入您的問題?補助\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
questionans
214.0中低收入醫療補助補助項目及標準為何？補助對象：設籍新竹市且符合下列情形之一者，於全民健康保險特約醫療院（所）就醫，未獲其他單位醫...
108.0申請假牙補助的資格及補助內容1.年滿65歲以上，經醫師評估缺牙需裝置活動假牙，且符合下列條件之一者：列冊低收入戶。領有中...
82.0特殊境遇家庭法律訴訟補助如何申請？補助額度如何？特殊境遇家庭之家庭暴力受害者無力負擔訴訟費用，得申請法律訴訟補助。其標準最高金額以新臺幣五萬...
\n", - "
" - ], - "text/plain": [ - " question \\\n", - "214.0 中低收入醫療補助補助項目及標準為何？ \n", - "108.0 申請假牙補助的資格及補助內容 \n", - "82.0 特殊境遇家庭法律訴訟補助如何申請？補助額度如何？ \n", - "\n", - " ans \n", - "214.0 補助對象：設籍新竹市且符合下列情形之一者，於全民健康保險特約醫療院（所）就醫，未獲其他單位醫... \n", - "108.0 1.年滿65歲以上，經醫師評估缺牙需裝置活動假牙，且符合下列條件之一者：列冊低收入戶。領有中... \n", - "82.0 特殊境遇家庭之家庭暴力受害者無力負擔訴訟費用，得申請法律訴訟補助。其標準最高金額以新臺幣五萬... " - ] - }, - "execution_count": 16, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "query = input('請輸入您的問題?')\n", - "retrieve(query)" - ] - }, - { - "cell_type": "code", - "execution_count": 115, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
questionans
100.0我已經年滿65歲領有國民年金老人年金及基本保證年金3628元，因家境清寒還可以再申請中低收入...不可以。所有的老人津貼，例如：國民年金、老農津貼、榮家院外就養金、中低收入老人生活津貼等只能...
111.0新竹市老人一般可領老人津貼6628元，該如何申請？新竹市一般老人領取國民年金基本保證年金3628元加上本市發放之安老津貼3000元，合計662...
321.0國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於...
\n", - "
" - ], - "text/plain": [ - " question \\\n", - "100.0 我已經年滿65歲領有國民年金老人年金及基本保證年金3628元，因家境清寒還可以再申請中低收入... \n", - "111.0 新竹市老人一般可領老人津貼6628元，該如何申請？ \n", - "321.0 國民年金保險被保險人如果是家庭收入較低者，國民年金保險費是否可以減免？補助標準為何？ \n", - "\n", - " ans \n", - "100.0 不可以。所有的老人津貼，例如：國民年金、老農津貼、榮家院外就養金、中低收入老人生活津貼等只能... \n", - "111.0 新竹市一般老人領取國民年金基本保證年金3628元加上本市發放之安老津貼3000元，合計662... \n", - "321.0 國民年金被保險人均可向戶籍所在地區公所社政課國民年金櫃台申請保費減免，全家人口平均月收入小於... " - ] - }, - "execution_count": 115, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "retrieve('老人年金')" - ] - }, - { - "cell_type": "code", - "execution_count": 113, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
questionans
15.0托育費用補助對象為何？補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總...
14.0托育費用補助標準為何？補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托...
19.0要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服...
\n", - "
" - ], - "text/plain": [ - " question \\\n", - "15.0 托育費用補助對象為何？ \n", - "14.0 托育費用補助標準為何？ \n", - "19.0 要申請托育補助一定要找加入居家托育服務中心的托育人員嗎？ \n", - "\n", - " ans \n", - "15.0 補助對象(一)一般家庭：父母(或監護人)雙方或單親一方經稅捐稽徵機關核定之最近1年綜合所得總... \n", - "14.0 補助標準如下（本項補助超過半個月、不滿1個月者以1個月計，未達半個月者以半個月計）：（一）托... \n", - "19.0 依照建構托育管理制度~托育費用補助申請須知，目前若家長要申請托育補助，寶寶必須給「居家托育服... " - ] - }, - "execution_count": 113, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "retrieve('托育')" - ] - }, - { - "cell_type": "code", - "execution_count": 114, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
questionans
204.0社會福利-急難救助核發救助對象？設籍本市之民眾，有下列情事之一者，得申請急難救助：一、戶內人口死亡無力殮葬。二、戶內人口遭受...
74.0遭遇特殊境遇家庭如何申請救助(申請方式)？申請人向戶籍所在地之區公所提出申請，區公所協助申請者辦理申請與初核後，符合條件者即層轉市府複...
203.0社會福利-我要到那裡申請急難救助？申請人應於事實發生之日起三個月內，向戶籍所在地區公所提出申請，但車資救助向社會處申請。東區區...
\n", - "
" - ], - "text/plain": [ - " question \\\n", - "204.0 社會福利-急難救助核發救助對象？ \n", - "74.0 遭遇特殊境遇家庭如何申請救助(申請方式)？ \n", - "203.0 社會福利-我要到那裡申請急難救助？ \n", - "\n", - " ans \n", - "204.0 設籍本市之民眾，有下列情事之一者，得申請急難救助：一、戶內人口死亡無力殮葬。二、戶內人口遭受... \n", - "74.0 申請人向戶籍所在地之區公所提出申請，區公所協助申請者辦理申請與初核後，符合條件者即層轉市府複... \n", - "203.0 申請人應於事實發生之日起三個月內，向戶籍所在地區公所提出申請，但車資救助向社會處申請。東區區... " - ] - }, - "execution_count": 114, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "retrieve('救助')" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "collapsed": true - }, - "outputs": [], - "source": [] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.6.4" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/09_IrImplementation/pic/DataFrame_Read.JPG b/09_IrImplementation/pic/DataFrame_Read.JPG deleted file mode 100644 index f52b123..0000000 Binary files a/09_IrImplementation/pic/DataFrame_Read.JPG and /dev/null differ diff --git a/09_IrImplementation/pic/TFIDF.JPG b/09_IrImplementation/pic/TFIDF.JPG deleted file mode 100644 index c1fb471..0000000 Binary files a/09_IrImplementation/pic/TFIDF.JPG and /dev/null differ diff --git a/09_IrImplementation/pic/build_termindex.JPG b/09_IrImplementation/pic/build_termindex.JPG deleted file mode 100644 index 6109207..0000000 Binary files a/09_IrImplementation/pic/build_termindex.JPG and /dev/null differ diff --git a/09_IrImplementation/pic/cosinesimilarity.JPG b/09_IrImplementation/pic/cosinesimilarity.JPG deleted file mode 100644 index a2499e6..0000000 Binary files a/09_IrImplementation/pic/cosinesimilarity.JPG and /dev/null differ diff --git a/09_IrImplementation/pic/doc2vec.JPG b/09_IrImplementation/pic/doc2vec.JPG deleted file mode 100644 index 7c76728..0000000 Binary files a/09_IrImplementation/pic/doc2vec.JPG and /dev/null differ diff --git a/09_IrImplementation/pic/preprocess.JPG b/09_IrImplementation/pic/preprocess.JPG deleted file mode 100644 index f809e5a..0000000 Binary files a/09_IrImplementation/pic/preprocess.JPG and /dev/null differ diff --git a/09_IrImplementation/pic/test1.JPG b/09_IrImplementation/pic/test1.JPG deleted file mode 100644 index c39e197..0000000 Binary files a/09_IrImplementation/pic/test1.JPG and /dev/null differ diff --git a/09_IrImplementation/pic/test2.JPG b/09_IrImplementation/pic/test2.JPG deleted file mode 100644 index 1c9cccb..0000000 Binary files a/09_IrImplementation/pic/test2.JPG and /dev/null differ diff --git a/09_IrImplementation/pic/test3.JPG b/09_IrImplementation/pic/test3.JPG deleted file mode 100644 index 36aa0d8..0000000 Binary files a/09_IrImplementation/pic/test3.JPG and /dev/null differ diff --git a/practice_generator.ipynb b/practice_generator.ipynb index e35e281..ad68ad1 100644 --- a/practice_generator.ipynb +++ b/practice_generator.ipynb @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "code", - "execution_count": 3, + "execution_count": 2, "metadata": {}, "outputs": [], "source": [ @@ -12,34 +12,35 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "dirs = [d for d in os.listdir() if bool(re.match('\\d\\_.+?', d))]\n", "for d in dirs: \n", - " if \"main.ipynb\" in os.listdir(d):\n", - " readfile_path = os.path.join(d, \"main.ipynb\")\n", - " writefile_path = os.path.join(d, \"practice.ipynb\")\n", - " readfile = open(readfile_path, 'r', encoding='utf8')\n", - " writefile = open(writefile_path, 'w', encoding='utf8')\n", - " starts_equal = False\n", - " starts_all = False\n", - " for line in readfile:\n", - " if \"#==============your works ends================#\" in line:\n", - " starts_equal = False\n", - " if \"#!==============your works ends================!#\" in line:\n", - " starts_all = False\n", - " if starts_equal and \"=\" in line:\n", - " line = line.split(\"=\")[0] + \"=\" + line[-5:]\n", - " if starts_all:\n", - " line = line[:5] + line[-5:]\n", - " if \"#=============your works starts===============#\" in line:\n", - " starts_equal = True\n", - " if \"#!=============your works starts===============!#\" in line:\n", - " starts_all = True\n", + " for f in os.listdir(d):\n", + " if f.startswith('main'):\n", + " readfile_path = os.path.join(d, f)\n", + " writefile_path = os.path.join(d, f.replace(\"main\", \"practice\"))\n", + " readfile = open(readfile_path, 'r', encoding='utf8')\n", + " writefile = open(writefile_path, 'w', encoding='utf8')\n", + " starts_equal = False\n", + " starts_all = False\n", + " for line in readfile:\n", + " if \"#==============your works ends================#\" in line:\n", + " starts_equal = False\n", + " if \"#!==============your works ends================!#\" in line:\n", + " starts_all = False\n", + " if starts_equal and \"=\" in line:\n", + " line = line.split(\"=\")[0] + \"=\" + line[-5:]\n", + " if starts_all:\n", + " line = line[:5] + line[-5:]\n", + " if \"#=============your works starts===============#\" in line:\n", + " starts_equal = True\n", + " if \"#!=============your works starts===============!#\" in line:\n", + " starts_all = True\n", "\n", - " writefile.write(line)\n", + " writefile.write(line)\n", " readfile.close()\n", " writefile.close()\n" ]