tensorflow/Credit Card Fraud Dectection using Keras (Imbalanced response).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Imbalanced classification: credit card fraud detection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train a classification model on data with highly imbalanced classes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "### 1. Read Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HEADER: \"Time\",\"V1\",\"V2\",\"V3\",\"V4\",\"V5\",\"V6\",\"V7\",\"V8\",\"V9\",\"V10\",\"V11\",\"V12\",\"V13\",\"V14\",\"V15\",\"V16\",\"V17\",\"V18\",\"V19\",\"V20\",\"V21\",\"V22\",\"V23\",\"V24\",\"V25\",\"V26\",\"V27\",\"V28\",\"Amount\",\"Class\"\n",
      "EXAMPLE FEATURES: [0.0, -1.3598071336738, -0.0727811733098497, 2.53634673796914, 1.37815522427443, -0.338320769942518, 0.462387777762292, 0.239598554061257, 0.0986979012610507, 0.363786969611213, 0.0907941719789316, -0.551599533260813, -0.617800855762348, -0.991389847235408, -0.311169353699879, 1.46817697209427, -0.470400525259478, 0.207971241929242, 0.0257905801985591, 0.403992960255733, 0.251412098239705, -0.018306777944153, 0.277837575558899, -0.110473910188767, 0.0669280749146731, 0.128539358273528, -0.189114843888824, 0.133558376740387, -0.0210530534538215, 149.62]\n",
      "features.shape: (284807, 30)\n",
      "targets.shape: (284807, 1)\n"
     ]
    }
   ],
   "source": [
    "import csv\n",
    "import numpy as np\n",
    "\n",
    "# Data from https://www.kaggle.com/mlg-ulb/creditcardfraud/\n",
    "# Download it to your local dir\n",
    "fname = \"C://Users//bokhy//Desktop//Python//Python-Projects//Pytorch//creditcard.csv\"\n",
    "\n",
    "all_features = []\n",
    "all_targets = []\n",
    "with open(fname) as f:\n",
    "    for i, line in enumerate(f):\n",
    "        if i == 0:\n",
    "            print(\"HEADER:\", line.strip())\n",
    "            continue  # Skip header\n",
    "        fields = line.strip().split(\",\")\n",
    "        all_features.append([float(v.replace('\"', \"\")) for v in fields[:-1]])\n",
    "        all_targets.append([int(fields[-1].replace('\"', \"\"))])\n",
    "        if i == 1:\n",
    "            print(\"EXAMPLE FEATURES:\", all_features[-1])\n",
    "\n",
    "features = np.array(all_features, dtype=\"float32\")\n",
    "targets = np.array(all_targets, dtype=\"uint8\")\n",
    "print(\"features.shape:\", features.shape)\n",
    "print(\"targets.shape:\", targets.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Train/Test Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of training samples: 227846\n",
      "Number of validation samples: 56961\n"
     ]
    }
   ],
   "source": [
    "num_val_samples = int(len(features) * 0.2)\n",
    "train_features = features[:-num_val_samples]\n",
    "train_targets = targets[:-num_val_samples]\n",
    "val_features = features[-num_val_samples:]\n",
    "val_targets = targets[-num_val_samples:]\n",
    "\n",
    "print(\"Number of training samples:\", len(train_features))\n",
    "print(\"Number of validation samples:\", len(val_features))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Analyze class imbalance in the Response variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of positive samples in training data: 417 (0.18% of total)\n"
     ]
    }
   ],
   "source": [
    "counts = np.bincount(train_targets[:, 0])\n",
    "print(\n",
    "    \"Number of positive samples in training data: {} ({:.2f}% of total)\".format(\n",
    "        counts[1], 100 * float(counts[1]) / len(train_targets)\n",
    "    )\n",
    ")\n",
    "\n",
    "weight_for_0 = 1.0 / counts[0]\n",
    "weight_for_1 = 1.0 / counts[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Normalize the data using training set statistics\n",
    "mean = np.mean(train_features, axis=0)\n",
    "train_features -= mean\n",
    "val_features -= mean\n",
    "std = np.std(train_features, axis=0)\n",
    "train_features /= std\n",
    "val_features /= std"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Create a Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense (Dense)                (None, 256)               7936      \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 256)               65792     \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 256)               65792     \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 1)                 257       \n",
      "=================================================================\n",
      "Total params: 139,777\n",
      "Trainable params: 139,777\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Binary classification model\n",
    "from tensorflow import keras\n",
    "\n",
    "model = keras.Sequential(\n",
    "    [\n",
    "        keras.layers.Dense(\n",
    "            256, activation=\"relu\", input_shape=(train_features.shape[-1],)\n",
    "        ),\n",
    "        keras.layers.Dense(256, activation=\"relu\"),\n",
    "        keras.layers.Dropout(0.3),\n",
    "        keras.layers.Dense(256, activation=\"relu\"),\n",
    "        keras.layers.Dropout(0.3),\n",
    "        keras.layers.Dense(1, activation=\"sigmoid\"),\n",
    "    ]\n",
    ")\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5. Model Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 227846 samples, validate on 56961 samples\n",
      "Epoch 1/30\n",
      "227846/227846 - 5s - loss: 6.2358e-07 - fn: 17.0000 - fp: 5062.0000 - tn: 222367.0000 - tp: 400.0000 - precision: 0.0732 - recall: 0.9592 - val_loss: 1.2858e-06 - val_fn: 8.0000 - val_fp: 621.0000 - val_tn: 56265.0000 - val_tp: 67.0000 - val_precision: 0.0974 - val_recall: 0.8933\n",
      "Epoch 2/30\n",
      "227846/227846 - 3s - loss: 5.2353e-07 - fn: 10.0000 - fp: 5086.0000 - tn: 222343.0000 - tp: 407.0000 - precision: 0.0741 - recall: 0.9760 - val_loss: 2.2980e-06 - val_fn: 9.0000 - val_fp: 543.0000 - val_tn: 56343.0000 - val_tp: 66.0000 - val_precision: 0.1084 - val_recall: 0.8800\n",
      "Epoch 3/30\n",
      "227846/227846 - 4s - loss: 1.0554e-06 - fn: 19.0000 - fp: 7910.0000 - tn: 219519.0000 - tp: 398.0000 - precision: 0.0479 - recall: 0.9544 - val_loss: 1.8638e-06 - val_fn: 5.0000 - val_fp: 4612.0000 - val_tn: 52274.0000 - val_tp: 70.0000 - val_precision: 0.0150 - val_recall: 0.9333\n",
      "Epoch 4/30\n",
      "227846/227846 - 5s - loss: 8.4623e-07 - fn: 19.0000 - fp: 5981.0000 - tn: 221448.0000 - tp: 398.0000 - precision: 0.0624 - recall: 0.9544 - val_loss: 1.1277e-06 - val_fn: 7.0000 - val_fp: 940.0000 - val_tn: 55946.0000 - val_tp: 68.0000 - val_precision: 0.0675 - val_recall: 0.9067\n",
      "Epoch 5/30\n",
      "227846/227846 - 4s - loss: 5.5524e-07 - fn: 10.0000 - fp: 5692.0000 - tn: 221737.0000 - tp: 407.0000 - precision: 0.0667 - recall: 0.9760 - val_loss: 1.3648e-06 - val_fn: 8.0000 - val_fp: 1332.0000 - val_tn: 55554.0000 - val_tp: 67.0000 - val_precision: 0.0479 - val_recall: 0.8933\n",
      "Epoch 6/30\n",
      "227846/227846 - 4s - loss: 4.9777e-07 - fn: 6.0000 - fp: 5501.0000 - tn: 221928.0000 - tp: 411.0000 - precision: 0.0695 - recall: 0.9856 - val_loss: 2.5905e-06 - val_fn: 7.0000 - val_fp: 997.0000 - val_tn: 55889.0000 - val_tp: 68.0000 - val_precision: 0.0638 - val_recall: 0.9067\n",
      "Epoch 7/30\n",
      "227846/227846 - 4s - loss: 5.0162e-07 - fn: 8.0000 - fp: 4970.0000 - tn: 222459.0000 - tp: 409.0000 - precision: 0.0760 - recall: 0.9808 - val_loss: 1.0967e-06 - val_fn: 6.0000 - val_fp: 2052.0000 - val_tn: 54834.0000 - val_tp: 69.0000 - val_precision: 0.0325 - val_recall: 0.9200\n",
      "Epoch 8/30\n",
      "227846/227846 - 4s - loss: 4.8823e-07 - fn: 7.0000 - fp: 5362.0000 - tn: 222067.0000 - tp: 410.0000 - precision: 0.0710 - recall: 0.9832 - val_loss: 1.9766e-06 - val_fn: 8.0000 - val_fp: 698.0000 - val_tn: 56188.0000 - val_tp: 67.0000 - val_precision: 0.0876 - val_recall: 0.8933\n",
      "Epoch 9/30\n",
      "227846/227846 - 4s - loss: 2.8699e-06 - fn: 14.0000 - fp: 8320.0000 - tn: 219109.0000 - tp: 403.0000 - precision: 0.0462 - recall: 0.9664 - val_loss: 3.2263e-06 - val_fn: 7.0000 - val_fp: 1953.0000 - val_tn: 54933.0000 - val_tp: 68.0000 - val_precision: 0.0336 - val_recall: 0.9067\n",
      "Epoch 10/30\n",
      "227846/227846 - 5s - loss: 3.5089e-06 - fn: 27.0000 - fp: 11682.0000 - tn: 215747.0000 - tp: 390.0000 - precision: 0.0323 - recall: 0.9353 - val_loss: 1.4504e-06 - val_fn: 10.0000 - val_fp: 480.0000 - val_tn: 56406.0000 - val_tp: 65.0000 - val_precision: 0.1193 - val_recall: 0.8667\n",
      "Epoch 11/30\n",
      "227846/227846 - 4s - loss: 1.0473e-06 - fn: 18.0000 - fp: 6770.0000 - tn: 220659.0000 - tp: 399.0000 - precision: 0.0557 - recall: 0.9568 - val_loss: 1.2955e-06 - val_fn: 8.0000 - val_fp: 1648.0000 - val_tn: 55238.0000 - val_tp: 67.0000 - val_precision: 0.0391 - val_recall: 0.8933\n",
      "Epoch 12/30\n",
      "227846/227846 - 4s - loss: 8.0038e-07 - fn: 11.0000 - fp: 5059.0000 - tn: 222370.0000 - tp: 406.0000 - precision: 0.0743 - recall: 0.9736 - val_loss: 1.7101e-06 - val_fn: 10.0000 - val_fp: 599.0000 - val_tn: 56287.0000 - val_tp: 65.0000 - val_precision: 0.0979 - val_recall: 0.8667\n",
      "Epoch 13/30\n",
      "227846/227846 - 4s - loss: 9.3132e-07 - fn: 8.0000 - fp: 5455.0000 - tn: 221974.0000 - tp: 409.0000 - precision: 0.0697 - recall: 0.9808 - val_loss: 2.4544e-06 - val_fn: 7.0000 - val_fp: 485.0000 - val_tn: 56401.0000 - val_tp: 68.0000 - val_precision: 0.1230 - val_recall: 0.9067\n",
      "Epoch 14/30\n",
      "227846/227846 - 5s - loss: 5.8621e-07 - fn: 8.0000 - fp: 3774.0000 - tn: 223655.0000 - tp: 409.0000 - precision: 0.0978 - recall: 0.9808 - val_loss: 3.1851e-06 - val_fn: 10.0000 - val_fp: 344.0000 - val_tn: 56542.0000 - val_tp: 65.0000 - val_precision: 0.1589 - val_recall: 0.8667\n",
      "Epoch 15/30\n",
      "227846/227846 - 5s - loss: 9.3993e-07 - fn: 11.0000 - fp: 8177.0000 - tn: 219252.0000 - tp: 406.0000 - precision: 0.0473 - recall: 0.9736 - val_loss: 2.8448e-06 - val_fn: 9.0000 - val_fp: 381.0000 - val_tn: 56505.0000 - val_tp: 66.0000 - val_precision: 0.1477 - val_recall: 0.8800\n",
      "Epoch 16/30\n",
      "227846/227846 - 4s - loss: 4.1142e-07 - fn: 7.0000 - fp: 3988.0000 - tn: 223441.0000 - tp: 410.0000 - precision: 0.0932 - recall: 0.9832 - val_loss: 2.5959e-06 - val_fn: 9.0000 - val_fp: 295.0000 - val_tn: 56591.0000 - val_tp: 66.0000 - val_precision: 0.1828 - val_recall: 0.8800\n",
      "Epoch 17/30\n",
      "227846/227846 - 4s - loss: 4.9894e-07 - fn: 7.0000 - fp: 3900.0000 - tn: 223529.0000 - tp: 410.0000 - precision: 0.0951 - recall: 0.9832 - val_loss: 2.5224e-06 - val_fn: 11.0000 - val_fp: 285.0000 - val_tn: 56601.0000 - val_tp: 64.0000 - val_precision: 0.1834 - val_recall: 0.8533\n",
      "Epoch 18/30\n",
      "227846/227846 - 4s - loss: 3.1725e-07 - fn: 3.0000 - fp: 3217.0000 - tn: 224212.0000 - tp: 414.0000 - precision: 0.1140 - recall: 0.9928 - val_loss: 3.5067e-06 - val_fn: 10.0000 - val_fp: 318.0000 - val_tn: 56568.0000 - val_tp: 65.0000 - val_precision: 0.1697 - val_recall: 0.8667\n",
      "Epoch 19/30\n",
      "227846/227846 - 4s - loss: 3.2463e-07 - fn: 3.0000 - fp: 3379.0000 - tn: 224050.0000 - tp: 414.0000 - precision: 0.1091 - recall: 0.9928 - val_loss: 4.3670e-06 - val_fn: 12.0000 - val_fp: 202.0000 - val_tn: 56684.0000 - val_tp: 63.0000 - val_precision: 0.2377 - val_recall: 0.8400\n",
      "Epoch 20/30\n",
      "227846/227846 - 4s - loss: 3.1807e-07 - fn: 4.0000 - fp: 3724.0000 - tn: 223705.0000 - tp: 413.0000 - precision: 0.0998 - recall: 0.9904 - val_loss: 3.6510e-06 - val_fn: 9.0000 - val_fp: 460.0000 - val_tn: 56426.0000 - val_tp: 66.0000 - val_precision: 0.1255 - val_recall: 0.8800\n",
      "Epoch 21/30\n",
      "227846/227846 - 4s - loss: 2.1827e-07 - fn: 1.0000 - fp: 2367.0000 - tn: 225062.0000 - tp: 416.0000 - precision: 0.1495 - recall: 0.9976 - val_loss: 4.3823e-06 - val_fn: 10.0000 - val_fp: 283.0000 - val_tn: 56603.0000 - val_tp: 65.0000 - val_precision: 0.1868 - val_recall: 0.8667\n",
      "Epoch 22/30\n",
      "227846/227846 - 4s - loss: 2.5676e-07 - fn: 3.0000 - fp: 2804.0000 - tn: 224625.0000 - tp: 414.0000 - precision: 0.1287 - recall: 0.9928 - val_loss: 4.4542e-06 - val_fn: 12.0000 - val_fp: 520.0000 - val_tn: 56366.0000 - val_tp: 63.0000 - val_precision: 0.1081 - val_recall: 0.8400\n",
      "Epoch 23/30\n",
      "227846/227846 - 4s - loss: 3.8872e-07 - fn: 2.0000 - fp: 4138.0000 - tn: 223291.0000 - tp: 415.0000 - precision: 0.0911 - recall: 0.9952 - val_loss: 5.5280e-06 - val_fn: 12.0000 - val_fp: 236.0000 - val_tn: 56650.0000 - val_tp: 63.0000 - val_precision: 0.2107 - val_recall: 0.8400\n",
      "Epoch 24/30\n",
      "227846/227846 - 5s - loss: 3.5192e-07 - fn: 3.0000 - fp: 3697.0000 - tn: 223732.0000 - tp: 414.0000 - precision: 0.1007 - recall: 0.9928 - val_loss: 4.9660e-06 - val_fn: 10.0000 - val_fp: 290.0000 - val_tn: 56596.0000 - val_tp: 65.0000 - val_precision: 0.1831 - val_recall: 0.8667\n",
      "Epoch 25/30\n",
      "227846/227846 - 4s - loss: 2.9852e-07 - fn: 3.0000 - fp: 3021.0000 - tn: 224408.0000 - tp: 414.0000 - precision: 0.1205 - recall: 0.9928 - val_loss: 5.5331e-06 - val_fn: 11.0000 - val_fp: 262.0000 - val_tn: 56624.0000 - val_tp: 64.0000 - val_precision: 0.1963 - val_recall: 0.8533\n",
      "Epoch 26/30\n",
      "227846/227846 - 4s - loss: 3.2259e-07 - fn: 4.0000 - fp: 3762.0000 - tn: 223667.0000 - tp: 413.0000 - precision: 0.0989 - recall: 0.9904 - val_loss: 6.4467e-06 - val_fn: 11.0000 - val_fp: 248.0000 - val_tn: 56638.0000 - val_tp: 64.0000 - val_precision: 0.2051 - val_recall: 0.8533\n",
      "Epoch 27/30\n",
      "227846/227846 - 4s - loss: 2.2511e-07 - fn: 1.0000 - fp: 2734.0000 - tn: 224695.0000 - tp: 416.0000 - precision: 0.1321 - recall: 0.9976 - val_loss: 5.9903e-06 - val_fn: 11.0000 - val_fp: 428.0000 - val_tn: 56458.0000 - val_tp: 64.0000 - val_precision: 0.1301 - val_recall: 0.8533\n",
      "Epoch 28/30\n",
      "227846/227846 - 4s - loss: 1.6367e-07 - fn: 2.0000 - fp: 1732.0000 - tn: 225697.0000 - tp: 415.0000 - precision: 0.1933 - recall: 0.9952 - val_loss: 5.0008e-06 - val_fn: 10.0000 - val_fp: 584.0000 - val_tn: 56302.0000 - val_tp: 65.0000 - val_precision: 0.1002 - val_recall: 0.8667\n",
      "Epoch 29/30\n",
      "227846/227846 - 4s - loss: 2.3816e-07 - fn: 2.0000 - fp: 2906.0000 - tn: 224523.0000 - tp: 415.0000 - precision: 0.1250 - recall: 0.9952 - val_loss: 5.0562e-06 - val_fn: 10.0000 - val_fp: 606.0000 - val_tn: 56280.0000 - val_tp: 65.0000 - val_precision: 0.0969 - val_recall: 0.8667\n",
      "Epoch 30/30\n",
      "227846/227846 - 4s - loss: 2.2919e-07 - fn: 2.0000 - fp: 2890.0000 - tn: 224539.0000 - tp: 415.0000 - precision: 0.1256 - recall: 0.9952 - val_loss: 6.1178e-06 - val_fn: 11.0000 - val_fp: 450.0000 - val_tn: 56436.0000 - val_tp: 64.0000 - val_precision: 0.1245 - val_recall: 0.8533\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x2db15471308>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Train the model with class_weight argument\n",
    "metrics = [\n",
    "    keras.metrics.FalseNegatives(name=\"fn\"),\n",
    "    keras.metrics.FalsePositives(name=\"fp\"),\n",
    "    keras.metrics.TrueNegatives(name=\"tn\"),\n",
    "    keras.metrics.TruePositives(name=\"tp\"),\n",
    "    keras.metrics.Precision(name=\"precision\"),\n",
    "    keras.metrics.Recall(name=\"recall\"),\n",
    "]\n",
    "\n",
    "model.compile(\n",
    "    optimizer=keras.optimizers.Adam(1e-2), loss=\"binary_crossentropy\", metrics=metrics\n",
    ")\n",
    "\n",
    "#save model for each epoch by making a callback\n",
    "#callbacks = [keras.callbacks.ModelCheckpoint(\"fraud_model_at_epoch_{epoch}.h5\")]\n",
    "\n",
    "class_weight = {0: weight_for_0, 1: weight_for_1}\n",
    "\n",
    "# Model fitting by adding test(validation)data\n",
    "model.fit(\n",
    "    train_features,\n",
    "    train_targets,\n",
    "    batch_size=2048,\n",
    "    epochs=30,\n",
    "    verbose=2,\n",
    "    #callbacks=callbacks,\n",
    "    validation_data=(val_features, val_targets),\n",
    "    class_weight=class_weight, ## important!\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "32/1 [================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================================] - 0s 779us/sample\n"
     ]
    },
    {
     "ename": "IndexError",
     "evalue": "list index out of range",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mIndexError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-13-669e86bb593d>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mloss\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0macc\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mevaluate\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mval_features\u001b[0m\u001b[1;33m)\u001b[0m  \u001b[1;31m# returns loss and metrics\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"loss: %.2f\"\u001b[0m \u001b[1;33m%\u001b[0m \u001b[0mloss\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"acc: %.2f\"\u001b[0m \u001b[1;33m%\u001b[0m \u001b[0macc\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\keras\\engine\\training.py\u001b[0m in \u001b[0;36mevaluate\u001b[1;34m(self, x, y, batch_size, verbose, sample_weight, steps, callbacks, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[0;32m    831\u001b[0m         \u001b[0mmax_queue_size\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mmax_queue_size\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    832\u001b[0m         \u001b[0mworkers\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mworkers\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 833\u001b[1;33m         use_multiprocessing=use_multiprocessing)\n\u001b[0m\u001b[0;32m    834\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    835\u001b[0m   def predict(self,\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\keras\\engine\\training_v2.py\u001b[0m in \u001b[0;36mevaluate\u001b[1;34m(self, model, x, y, batch_size, verbose, sample_weight, steps, callbacks, **kwargs)\u001b[0m\n\u001b[0;32m    454\u001b[0m     return self._model_iteration(\n\u001b[0;32m    455\u001b[0m         \u001b[0mmodel\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mModeKeys\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mTEST\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mx\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mverbose\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 456\u001b[1;33m         sample_weight=sample_weight, steps=steps, callbacks=callbacks, **kwargs)\n\u001b[0m\u001b[0;32m    457\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    458\u001b[0m   def predict(self, model, x, batch_size=None, verbose=0, steps=None,\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\keras\\engine\\training_v2.py\u001b[0m in \u001b[0;36m_model_iteration\u001b[1;34m(self, model, mode, x, y, batch_size, verbose, sample_weight, steps, callbacks, **kwargs)\u001b[0m\n\u001b[0;32m    442\u001b[0m               \u001b[0mmode\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mmode\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    443\u001b[0m               \u001b[0mtraining_context\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mtraining_context\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 444\u001b[1;33m               total_epochs=1)\n\u001b[0m\u001b[0;32m    445\u001b[0m           \u001b[0mcbks\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmake_logs\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mepoch_logs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mresult\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmode\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    446\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\keras\\engine\\training_v2.py\u001b[0m in \u001b[0;36mrun_one_epoch\u001b[1;34m(model, iterator, execution_function, dataset_size, batch_size, strategy, steps_per_epoch, num_samples, mode, training_context, total_epochs)\u001b[0m\n\u001b[0;32m    121\u001b[0m         step=step, mode=mode, size=current_batch_size) as batch_logs:\n\u001b[0;32m    122\u001b[0m       \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 123\u001b[1;33m         \u001b[0mbatch_outs\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mexecution_function\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    124\u001b[0m       \u001b[1;32mexcept\u001b[0m \u001b[1;33m(\u001b[0m\u001b[0mStopIteration\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0merrors\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mOutOfRangeError\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    125\u001b[0m         \u001b[1;31m# TODO(kaftan): File bug about tf function and errors.OutOfRangeError?\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\keras\\engine\\training_v2_utils.py\u001b[0m in \u001b[0;36mexecution_function\u001b[1;34m(input_fn)\u001b[0m\n\u001b[0;32m     84\u001b[0m     \u001b[1;31m# `numpy` translates Tensors to values in Eager mode.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     85\u001b[0m     return nest.map_structure(_non_none_constant_value,\n\u001b[1;32m---> 86\u001b[1;33m                               distributed_function(input_fn))\n\u001b[0m\u001b[0;32m     87\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     88\u001b[0m   \u001b[1;32mreturn\u001b[0m \u001b[0mexecution_function\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\eager\\def_function.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, *args, **kwds)\u001b[0m\n\u001b[0;32m    455\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    456\u001b[0m     \u001b[0mtracing_count\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_get_tracing_count\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 457\u001b[1;33m     \u001b[0mresult\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    458\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mtracing_count\u001b[0m \u001b[1;33m==\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_get_tracing_count\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    459\u001b[0m       \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_call_counter\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcalled_without_tracing\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\eager\\def_function.py\u001b[0m in \u001b[0;36m_call\u001b[1;34m(self, *args, **kwds)\u001b[0m\n\u001b[0;32m    492\u001b[0m       \u001b[1;31m# In this case we have not created variables on the first call. So we can\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    493\u001b[0m       \u001b[1;31m# run the first trace but we should fail if variables are created.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 494\u001b[1;33m       \u001b[0mresults\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_stateful_fn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    495\u001b[0m       \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_created_variables\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    496\u001b[0m         raise ValueError(\"Creating variables on a non-first call to a function\"\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\eager\\function.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1820\u001b[0m   \u001b[1;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1821\u001b[0m     \u001b[1;34m\"\"\"Calls a graph function specialized to the inputs.\"\"\"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1822\u001b[1;33m     \u001b[0mgraph_function\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkwargs\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_maybe_define_function\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1823\u001b[0m     \u001b[1;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_filtered_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m  \u001b[1;31m# pylint: disable=protected-access\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1824\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\eager\\function.py\u001b[0m in \u001b[0;36m_maybe_define_function\u001b[1;34m(self, args, kwargs)\u001b[0m\n\u001b[0;32m   2148\u001b[0m         \u001b[0mgraph_function\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mprimary\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcache_key\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2149\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mgraph_function\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2150\u001b[1;33m           \u001b[0mgraph_function\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_create_graph_function\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2151\u001b[0m           \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mprimary\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mcache_key\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2152\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\eager\\function.py\u001b[0m in \u001b[0;36m_create_graph_function\u001b[1;34m(self, args, kwargs, override_flat_arg_shapes)\u001b[0m\n\u001b[0;32m   2039\u001b[0m             \u001b[0marg_names\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0marg_names\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2040\u001b[0m             \u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2041\u001b[1;33m             capture_by_value=self._capture_by_value),\n\u001b[0m\u001b[0;32m   2042\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_function_attributes\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2043\u001b[0m         \u001b[1;31m# Tell the ConcreteFunction to clean up its graph once it goes out of\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\framework\\func_graph.py\u001b[0m in \u001b[0;36mfunc_graph_from_py_func\u001b[1;34m(name, python_func, args, kwargs, signature, func_graph, autograph, autograph_options, add_control_dependencies, arg_names, op_return_value, collections, capture_by_value, override_flat_arg_shapes)\u001b[0m\n\u001b[0;32m    913\u001b[0m                                           converted_func)\n\u001b[0;32m    914\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 915\u001b[1;33m       \u001b[0mfunc_outputs\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mpython_func\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0mfunc_args\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mfunc_kwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    916\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    917\u001b[0m       \u001b[1;31m# invariant: `func_outputs` contains only Tensors, CompositeTensors,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\eager\\def_function.py\u001b[0m in \u001b[0;36mwrapped_fn\u001b[1;34m(*args, **kwds)\u001b[0m\n\u001b[0;32m    356\u001b[0m         \u001b[1;31m# __wrapped__ allows AutoGraph to swap in a converted function. We give\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    357\u001b[0m         \u001b[1;31m# the function a weak reference to itself to avoid a reference cycle.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 358\u001b[1;33m         \u001b[1;32mreturn\u001b[0m \u001b[0mweak_wrapped_fn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__wrapped__\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    359\u001b[0m     \u001b[0mweak_wrapped_fn\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mweakref\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mref\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mwrapped_fn\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    360\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\keras\\engine\\training_v2_utils.py\u001b[0m in \u001b[0;36mdistributed_function\u001b[1;34m(input_iterator)\u001b[0m\n\u001b[0;32m     71\u001b[0m     \u001b[0mstrategy\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mdistribution_strategy_context\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_strategy\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     72\u001b[0m     outputs = strategy.experimental_run_v2(\n\u001b[1;32m---> 73\u001b[1;33m         per_replica_function, args=(model, x, y, sample_weights))\n\u001b[0m\u001b[0;32m     74\u001b[0m     \u001b[1;31m# Out of PerReplica outputs reduce or pick values to return.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     75\u001b[0m     all_outputs = dist_utils.unwrap_output_dict(\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\distribute\\distribute_lib.py\u001b[0m in \u001b[0;36mexperimental_run_v2\u001b[1;34m(self, fn, args, kwargs)\u001b[0m\n\u001b[0;32m    758\u001b[0m       fn = autograph.tf_convert(fn, ag_ctx.control_status_ctx(),\n\u001b[0;32m    759\u001b[0m                                 convert_by_default=False)\n\u001b[1;32m--> 760\u001b[1;33m       \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_extended\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcall_for_each_replica\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margs\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    761\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    762\u001b[0m   \u001b[1;32mdef\u001b[0m \u001b[0mreduce\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreduce_op\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\distribute\\distribute_lib.py\u001b[0m in \u001b[0;36mcall_for_each_replica\u001b[1;34m(self, fn, args, kwargs)\u001b[0m\n\u001b[0;32m   1785\u001b[0m       \u001b[0mkwargs\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m{\u001b[0m\u001b[1;33m}\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1786\u001b[0m     \u001b[1;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_container_strategy\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mscope\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1787\u001b[1;33m       \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_call_for_each_replica\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1788\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1789\u001b[0m   \u001b[1;32mdef\u001b[0m \u001b[0m_call_for_each_replica\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\distribute\\distribute_lib.py\u001b[0m in \u001b[0;36m_call_for_each_replica\u001b[1;34m(self, fn, args, kwargs)\u001b[0m\n\u001b[0;32m   2130\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_container_strategy\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2131\u001b[0m         replica_id_in_sync_group=constant_op.constant(0, dtypes.int32)):\n\u001b[1;32m-> 2132\u001b[1;33m       \u001b[1;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2133\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2134\u001b[0m   \u001b[1;32mdef\u001b[0m \u001b[0m_reduce_to\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreduce_op\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdestinations\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\autograph\\impl\\api.py\u001b[0m in \u001b[0;36mwrapper\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m    290\u001b[0m   \u001b[1;32mdef\u001b[0m \u001b[0mwrapper\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    291\u001b[0m     \u001b[1;32mwith\u001b[0m \u001b[0mag_ctx\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mControlStatusCtx\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mstatus\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mag_ctx\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mStatus\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mDISABLED\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 292\u001b[1;33m       \u001b[1;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    293\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    294\u001b[0m   \u001b[1;32mif\u001b[0m \u001b[0minspect\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0misfunction\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mor\u001b[0m \u001b[0minspect\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mismethod\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\keras\\engine\\training_v2_utils.py\u001b[0m in \u001b[0;36mtest_on_batch\u001b[1;34m(model, x, y, sample_weight, reset_metrics)\u001b[0m\n\u001b[0;32m    325\u001b[0m       \u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    326\u001b[0m       \u001b[0msample_weights\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0msample_weights\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 327\u001b[1;33m       output_loss_metrics=model._output_loss_metrics)\n\u001b[0m\u001b[0;32m    328\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    329\u001b[0m   \u001b[1;32mif\u001b[0m \u001b[0mreset_metrics\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\keras\\engine\\training_eager.py\u001b[0m in \u001b[0;36mtest_on_batch\u001b[1;34m(model, inputs, targets, sample_weights, output_loss_metrics)\u001b[0m\n\u001b[0;32m    352\u001b[0m             \u001b[0msample_weights\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0msample_weights\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    353\u001b[0m             \u001b[0mtraining\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mFalse\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 354\u001b[1;33m             output_loss_metrics=output_loss_metrics))\n\u001b[0m\u001b[0;32m    355\u001b[0m   \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    356\u001b[0m     \u001b[0mouts\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[0mouts\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Roaming\\Python\\Python37\\site-packages\\tensorflow_core\\python\\keras\\engine\\training_eager.py\u001b[0m in \u001b[0;36m_model_loss\u001b[1;34m(model, inputs, targets, output_loss_metrics, sample_weights, training)\u001b[0m\n\u001b[0;32m    164\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    165\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mloss_fn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'reduction'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 166\u001b[1;33m           \u001b[0mper_sample_losses\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mloss_fn\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcall\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtargets\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mouts\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    167\u001b[0m           weighted_losses = losses_utils.compute_weighted_loss(\n\u001b[0;32m    168\u001b[0m               \u001b[0mper_sample_losses\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mIndexError\u001b[0m: list index out of range"
     ]
    }
   ],
   "source": [
    "loss, acc = model.evaluate(val_features)  # returns loss and metrics\n",
    "print(\"loss: %.2f\" % loss)\n",
    "print(\"acc: %.2f\" % acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusions\n",
    "## At the end of training, out of 56,961 validation transactions, we are:\n",
    "\n",
    "- >Correctly identifying 64 of them as fraudulent\n",
    "- >Missing 11 fraudulent transactions at the cost of incorrectly flagging 450 legitimate transactions"
   ]
  },
  {
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}