Tugas_6_save_model&plot.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Tugas_6_save-model&plot.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/MrezaFd/Tugas-Besar-ML-059-055-/blob/main/Tugas_6_save_model%26plot.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pg3l1ejwCF9P"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 94
        },
        "id": "_yhI15_dBJI1",
        "outputId": "604780a0-ddb4-4c87-d466-96464d2a895d"
      },
      "source": [
        "from google.colab import files #import lib\n",
        "files.upload()"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-96561d8a-2774-4bf7-b926-894247cded5a\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-96561d8a-2774-4bf7-b926-894247cded5a\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving kaggle.json to kaggle.json\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'kaggle.json': b'{\"username\":\"hexazor\",\"key\":\"663a14597732196faa9d3257dd4ff4df\"}'}"
            ]
          },
          "metadata": {},
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T-PttF09A1jP"
      },
      "source": [
        "!mkdir -p ~/.kaggle #membuat directory bernama kaggle \n",
        "!cp kaggle.json ~/.kaggle/ #copy file kaggle.json upload tadi ke directory baru\n",
        "!chmod 600 ~/.kaggle/kaggle.json #berikan izin untuk file kaggle.jsonv"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OvexdoKwBZ5V",
        "outputId": "f72120ca-a5ad-460d-cbd2-b250563b82d8"
      },
      "source": [
        "#Unduh dataset\n",
        "!kaggle datasets download -d fanconic/skin-cancer-malignant-vs-benign"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading skin-cancer-malignant-vs-benign.zip to /content\n",
            "100% 324M/325M [00:02<00:00, 148MB/s]\n",
            "100% 325M/325M [00:02<00:00, 152MB/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4rtnNrFsEDIf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7cb9c2f1-762d-4723-ba71-8b44719eebce"
      },
      "source": [
        "from zipfile import ZipFile\n",
        "file_name = \"skin-cancer-malignant-vs-benign.zip\"\n",
        "\n",
        "with ZipFile(file_name,'r') as zip :\n",
        "  zip.extractall()\n",
        "  print('Berhasil')"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Berhasil\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "31Pr3YSaDHnn"
      },
      "source": [
        "!mkdir a"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u6Wd9eg8DOao"
      },
      "source": [
        "!mv -t a test train"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y9L_1kqnEEjc"
      },
      "source": [
        "# Definisikan path untuk data train dan data validation\n",
        "import os\n",
        "base_dir = '/content/a'\n",
        "\n",
        "\n",
        "train_dir = os.path.join(base_dir, 'train')\n",
        "validation_dir = os.path.join(base_dir, 'test')"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9gc0BCSREF0M"
      },
      "source": [
        "import cv2\n",
        "import numpy as np\n",
        "\n",
        "# Gather data train\n",
        "train_data = []\n",
        "train_label = []\n",
        "for r, d, f in os.walk(train_dir):\n",
        "    for file in f:\n",
        "        if \".jpg\" in file:\n",
        "            imagePath = os.path.join(r, file)\n",
        "            image = cv2.imread(imagePath)\n",
        "            image = cv2.resize(image, (150,150))\n",
        "            train_data.append(image)\n",
        "            label = imagePath.split(os.path.sep)[-2]\n",
        "            train_label.append(label)\n",
        "\n",
        "train_data = np.array(train_data)\n",
        "train_label = np.array(train_label)"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FuoKFhhVEG_O"
      },
      "source": [
        "# Gather data validation\n",
        "val_data = []\n",
        "val_label = []\n",
        "for r, d, f in os.walk(validation_dir):\n",
        "    for file in f:\n",
        "        if \".jpg\" in file:\n",
        "            imagePath = os.path.join(r, file)\n",
        "            image = cv2.imread(imagePath)\n",
        "            image = cv2.resize(image, (150,150))\n",
        "            val_data.append(image)\n",
        "            label = imagePath.split(os.path.sep)[-2]\n",
        "            val_label.append(label)\n",
        "\n",
        "val_data = np.array(val_data)\n",
        "val_label = np.array(val_label)"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kiXbwrN3ELqv"
      },
      "source": [
        "<h3>Data Preprocessing</h3>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m5VhnvtrEI1I",
        "outputId": "f0c091f0-42cc-4cbf-ec65-d5332aa1e375"
      },
      "source": [
        "# Tampilkan shape dari data train dan data validation\n",
        "print(\"Train Data = \", train_data.shape)\n",
        "print(\"Train Label = \", train_label.shape)\n",
        "print(\"Validation Data = \", val_data.shape)\n",
        "print(\"Validation Label = \", val_label.shape)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Data =  (2637, 150, 150, 3)\n",
            "Train Label =  (2637,)\n",
            "Validation Data =  (660, 150, 150, 3)\n",
            "Validation Label =  (660,)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q1LVp6dCENne",
        "outputId": "5cebc8e8-d1e9-4e04-a8fb-2128f0eb0889"
      },
      "source": [
        "# Normalisasi dataset\n",
        "print(\"Data sebelum di-normalisasi \", train_data[0][0][0])\n",
        "\n",
        "x_train = train_data.astype('float32') / 255.0\n",
        "x_val = val_data.astype('float32') / 255.0\n",
        "print(\"Data setelah di-normalisasi \", x_train[0][0][0])"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Data sebelum di-normalisasi  [ 88 101 145]\n",
            "Data setelah di-normalisasi  [0.34509805 0.39607844 0.5686275 ]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tc7FivC6EPhm",
        "outputId": "2dad7ece-1e27-437d-948a-48c2a554445b"
      },
      "source": [
        "# Transformasi label encoder\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "\n",
        "print(\"Label sebelum di-encoder \", train_label[995:2005])\n",
        "\n",
        "lb = LabelEncoder()\n",
        "y_train = lb.fit_transform(train_label)\n",
        "y_val = lb.fit_transform(val_label)\n",
        "\n",
        "print(\"Label setelah di-encoder \", y_train[995:2005])"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Label sebelum di-encoder  ['malignant' 'malignant' 'malignant' ... 'benign' 'benign' 'benign']\n",
            "Label setelah di-encoder  [1 1 1 ... 0 0 0]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4jVWbOhOEzQJ"
      },
      "source": [
        "<h3>Definisikan Model</h3>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "usLiFiwAEvVb"
      },
      "source": [
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import InputLayer, Dense, Conv2D, MaxPool2D, Flatten\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint\n",
        "import tensorflow as tf"
      ],
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rDwzxacpE1ei"
      },
      "source": [
        "# Feature Extraction Layer\n",
        "model = Sequential()\n",
        "\n",
        "model.add(InputLayer(input_shape=[150,150,3]))\n",
        "model.add(Conv2D(filters=16, kernel_size=3, strides=1, padding='same', activation='relu'))\n",
        "model.add(MaxPool2D(pool_size=2, padding='same'))\n",
        "model.add(Conv2D(filters=32, kernel_size=3, strides=1, padding='same', activation='relu'))\n",
        "model.add(MaxPool2D(pool_size=2, padding='same'))\n",
        "model.add(Conv2D(filters=64, kernel_size=3, strides=1, padding='same', activation='relu'))\n",
        "model.add(MaxPool2D(pool_size=2, padding='same'))\n",
        "model.add(Flatten())"
      ],
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Sud1AUF0E2k8"
      },
      "source": [
        "# Fully Connected Layer\n",
        "model.add(Dense(128, activation='relu'))\n",
        "model.add(Dense(1, activation='sigmoid'))"
      ],
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XW-KE9ePE466",
        "outputId": "3aef711a-51b6-4ace-9280-a41f0bd66495"
      },
      "source": [
        "# Print model summary\n",
        "print(model.summary())"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_6 (Conv2D)           (None, 150, 150, 16)      448       \n",
            "                                                                 \n",
            " max_pooling2d_6 (MaxPooling  (None, 75, 75, 16)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_7 (Conv2D)           (None, 75, 75, 32)        4640      \n",
            "                                                                 \n",
            " max_pooling2d_7 (MaxPooling  (None, 38, 38, 32)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_8 (Conv2D)           (None, 38, 38, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_8 (MaxPooling  (None, 19, 19, 64)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten_2 (Flatten)         (None, 23104)             0         \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 128)               2957440   \n",
            "                                                                 \n",
            " dense_5 (Dense)             (None, 1)                 129       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 2,981,153\n",
            "Trainable params: 2,981,153\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MbyolzG3JgvR"
      },
      "source": [
        "best_model_cancer_path = 'model-cancer.h5'\n",
        "checkpoint_callback = ModelCheckpoint(best_model_cancer_path,\n",
        "                                     monitor='val_acc',\n",
        "                                     save_best_only=True,\n",
        "                                     verbose=1)\n",
        "callbacks_list = [checkpoint_callback]"
      ],
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Tais64D7E5Zo",
        "outputId": "02f2017f-a2ee-4693-90cb-fa0176520024"
      },
      "source": [
        "# Compile model\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "\n",
        "model.compile(optimizer=Adam(lr=0.001), \n",
        "              loss='binary_crossentropy', \n",
        "              metrics=['acc'])"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NTNBH19iE8kt"
      },
      "source": [
        "<h3>Training Model</h3>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "p4Vr8dyvE6pM",
        "outputId": "8f419a9e-1ad5-4f0c-ab27-d2d9b7b805c1"
      },
      "source": [
        "H = model.fit(x_train, y_train, batch_size=20, epochs=25, validation_data=(x_val, y_val),callbacks=[callbacks_list])\n"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/25\n",
            "132/132 [==============================] - ETA: 0s - loss: 0.6077 - acc: 0.6701\n",
            "Epoch 00001: val_acc improved from -inf to 0.78030, saving model to model-cancer.h5\n",
            "132/132 [==============================] - 4s 28ms/step - loss: 0.6077 - acc: 0.6701 - val_loss: 0.4337 - val_acc: 0.7803\n",
            "Epoch 2/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.4356 - acc: 0.7824\n",
            "Epoch 00002: val_acc improved from 0.78030 to 0.82727, saving model to model-cancer.h5\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 0.4352 - acc: 0.7827 - val_loss: 0.3581 - val_acc: 0.8273\n",
            "Epoch 3/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.3792 - acc: 0.8198\n",
            "Epoch 00003: val_acc improved from 0.82727 to 0.82879, saving model to model-cancer.h5\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.3788 - acc: 0.8199 - val_loss: 0.3478 - val_acc: 0.8288\n",
            "Epoch 4/25\n",
            "130/132 [============================>.] - ETA: 0s - loss: 0.3627 - acc: 0.8319\n",
            "Epoch 00004: val_acc did not improve from 0.82879\n",
            "132/132 [==============================] - 3s 23ms/step - loss: 0.3667 - acc: 0.8294 - val_loss: 0.4324 - val_acc: 0.8015\n",
            "Epoch 5/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.3761 - acc: 0.8118\n",
            "Epoch 00005: val_acc improved from 0.82879 to 0.83636, saving model to model-cancer.h5\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.3756 - acc: 0.8115 - val_loss: 0.3473 - val_acc: 0.8364\n",
            "Epoch 6/25\n",
            "132/132 [==============================] - ETA: 0s - loss: 0.3190 - acc: 0.8419\n",
            "Epoch 00006: val_acc did not improve from 0.83636\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.3190 - acc: 0.8419 - val_loss: 0.3450 - val_acc: 0.8242\n",
            "Epoch 7/25\n",
            "132/132 [==============================] - ETA: 0s - loss: 0.3039 - acc: 0.8548\n",
            "Epoch 00007: val_acc did not improve from 0.83636\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.3039 - acc: 0.8548 - val_loss: 0.4012 - val_acc: 0.8121\n",
            "Epoch 8/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.3277 - acc: 0.8450\n",
            "Epoch 00008: val_acc did not improve from 0.83636\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.3283 - acc: 0.8445 - val_loss: 0.4150 - val_acc: 0.7939\n",
            "Epoch 9/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.3016 - acc: 0.8573\n",
            "Epoch 00009: val_acc did not improve from 0.83636\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 0.3011 - acc: 0.8582 - val_loss: 0.3645 - val_acc: 0.8242\n",
            "Epoch 10/25\n",
            "132/132 [==============================] - ETA: 0s - loss: 0.2658 - acc: 0.8752\n",
            "Epoch 00010: val_acc improved from 0.83636 to 0.84091, saving model to model-cancer.h5\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 0.2658 - acc: 0.8752 - val_loss: 0.3603 - val_acc: 0.8409\n",
            "Epoch 11/25\n",
            "132/132 [==============================] - ETA: 0s - loss: 0.2447 - acc: 0.8889\n",
            "Epoch 00011: val_acc did not improve from 0.84091\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.2447 - acc: 0.8889 - val_loss: 0.3799 - val_acc: 0.8212\n",
            "Epoch 12/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.2341 - acc: 0.8954\n",
            "Epoch 00012: val_acc did not improve from 0.84091\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.2334 - acc: 0.8957 - val_loss: 0.3841 - val_acc: 0.8227\n",
            "Epoch 13/25\n",
            "132/132 [==============================] - ETA: 0s - loss: 0.1940 - acc: 0.9124\n",
            "Epoch 00013: val_acc did not improve from 0.84091\n",
            "132/132 [==============================] - 3s 23ms/step - loss: 0.1940 - acc: 0.9124 - val_loss: 0.4546 - val_acc: 0.8197\n",
            "Epoch 14/25\n",
            "130/132 [============================>.] - ETA: 0s - loss: 0.1749 - acc: 0.9277\n",
            "Epoch 00014: val_acc did not improve from 0.84091\n",
            "132/132 [==============================] - 3s 23ms/step - loss: 0.1744 - acc: 0.9283 - val_loss: 0.4269 - val_acc: 0.8333\n",
            "Epoch 15/25\n",
            "130/132 [============================>.] - ETA: 0s - loss: 0.1806 - acc: 0.9246\n",
            "Epoch 00015: val_acc did not improve from 0.84091\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.1804 - acc: 0.9245 - val_loss: 0.4561 - val_acc: 0.8106\n",
            "Epoch 16/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.1210 - acc: 0.9557\n",
            "Epoch 00016: val_acc did not improve from 0.84091\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.1209 - acc: 0.9556 - val_loss: 0.5839 - val_acc: 0.8273\n",
            "Epoch 17/25\n",
            "130/132 [============================>.] - ETA: 0s - loss: 0.1016 - acc: 0.9662\n",
            "Epoch 00017: val_acc did not improve from 0.84091\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.1026 - acc: 0.9655 - val_loss: 0.5852 - val_acc: 0.8242\n",
            "Epoch 18/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.1161 - acc: 0.9542\n",
            "Epoch 00018: val_acc did not improve from 0.84091\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.1163 - acc: 0.9541 - val_loss: 0.6296 - val_acc: 0.8152\n",
            "Epoch 19/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.0714 - acc: 0.9756\n",
            "Epoch 00019: val_acc did not improve from 0.84091\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.0720 - acc: 0.9754 - val_loss: 0.7064 - val_acc: 0.8394\n",
            "Epoch 20/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.0553 - acc: 0.9802\n",
            "Epoch 00020: val_acc improved from 0.84091 to 0.84848, saving model to model-cancer.h5\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 0.0550 - acc: 0.9803 - val_loss: 0.6793 - val_acc: 0.8485\n",
            "Epoch 21/25\n",
            "132/132 [==============================] - ETA: 0s - loss: 0.0535 - acc: 0.9818\n",
            "Epoch 00021: val_acc did not improve from 0.84848\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.0535 - acc: 0.9818 - val_loss: 0.7584 - val_acc: 0.7924\n",
            "Epoch 22/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.0721 - acc: 0.9767\n",
            "Epoch 00022: val_acc did not improve from 0.84848\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.0720 - acc: 0.9769 - val_loss: 0.7658 - val_acc: 0.8106\n",
            "Epoch 23/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.0264 - acc: 0.9920\n",
            "Epoch 00023: val_acc did not improve from 0.84848\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.0262 - acc: 0.9920 - val_loss: 0.8251 - val_acc: 0.8455\n",
            "Epoch 24/25\n",
            "131/132 [============================>.] - ETA: 0s - loss: 0.0686 - acc: 0.9790\n",
            "Epoch 00024: val_acc did not improve from 0.84848\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.0682 - acc: 0.9791 - val_loss: 0.7343 - val_acc: 0.8303\n",
            "Epoch 25/25\n",
            "132/132 [==============================] - ETA: 0s - loss: 0.0461 - acc: 0.9837\n",
            "Epoch 00025: val_acc did not improve from 0.84848\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.0461 - acc: 0.9837 - val_loss: 0.8348 - val_acc: 0.8258\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9j4CkHMQI8dw"
      },
      "source": [
        "# **Model Save Weight**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mehtsfHKJBd0"
      },
      "source": [
        "model.save_weights('model-cancer.h5')"
      ],
      "execution_count": 47,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AeDTsyJ0I7wN",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "70a7ba0d-5e82-4c1e-89a7-129b2bc9a46e"
      },
      "source": [
        "model.get_weights()"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[array([[[[-0.01698263, -0.18792966, -0.14404342, -0.12046345,\n",
              "            0.0078126 ,  0.04350316, -0.04448006,  0.04460702,\n",
              "            0.20817134, -0.019699  ,  0.0304031 ,  0.09878087,\n",
              "           -0.19677885, -0.08543588, -0.06177767,  0.13678588],\n",
              "          [ 0.14398986, -0.02507931,  0.04367113, -0.09817869,\n",
              "            0.12227573, -0.1978103 ,  0.00617655,  0.05821764,\n",
              "           -0.08795018,  0.19315298, -0.04679071, -0.01559609,\n",
              "           -0.07561266, -0.1527082 ,  0.05750867,  0.03327572],\n",
              "          [ 0.139741  ,  0.07425105,  0.17406656,  0.12900797,\n",
              "           -0.10166358, -0.00222063,  0.03752314, -0.20128404,\n",
              "            0.10502548, -0.10467149,  0.03805979, -0.10273945,\n",
              "            0.16849871,  0.05025603, -0.09581448,  0.07413822]],\n",
              " \n",
              "         [[-0.0394084 , -0.32348967,  0.10921156,  0.04081375,\n",
              "            0.05872468,  0.08332881, -0.00969216,  0.11017407,\n",
              "           -0.19890597, -0.16480377,  0.14204735, -0.04710543,\n",
              "           -0.19650355,  0.12557672, -0.08334873, -0.13669498],\n",
              "          [-0.08911955, -0.02660539,  0.09656943,  0.03780296,\n",
              "            0.01750374,  0.03741892,  0.02637054,  0.16260333,\n",
              "           -0.1891342 ,  0.15027621,  0.12146039,  0.08685737,\n",
              "           -0.20524876,  0.10348427,  0.07112826, -0.02406617],\n",
              "          [-0.22101463,  0.06548027, -0.09660815, -0.11295629,\n",
              "           -0.05738249,  0.04516503,  0.1609366 , -0.10210988,\n",
              "            0.11206035, -0.07214146, -0.2115061 ,  0.07279816,\n",
              "            0.1976068 ,  0.06869777, -0.21773843, -0.08313154]],\n",
              " \n",
              "         [[-0.00085182, -0.10725388,  0.05339447,  0.1748469 ,\n",
              "           -0.07604785,  0.07097037, -0.04006259, -0.03417918,\n",
              "           -0.24399954, -0.04489437, -0.15091428,  0.11680367,\n",
              "           -0.12994134, -0.19214383,  0.18843582,  0.11354497],\n",
              "          [ 0.09941281,  0.23881386, -0.17914556,  0.14690652,\n",
              "           -0.12859306, -0.07010537, -0.18370293,  0.15762888,\n",
              "            0.02891432,  0.23090054, -0.03295344,  0.13938305,\n",
              "            0.09297816, -0.17130606,  0.19847986, -0.10050391],\n",
              "          [ 0.01001362, -0.06086392, -0.14803213, -0.07829326,\n",
              "           -0.0324956 ,  0.10955886, -0.16348031,  0.02769195,\n",
              "           -0.03306177, -0.01280448, -0.11936536, -0.19705367,\n",
              "            0.13876283,  0.19717288,  0.17687032, -0.02816627]]],\n",
              " \n",
              " \n",
              "        [[[ 0.0493806 , -0.31572387,  0.1426604 , -0.09906369,\n",
              "           -0.17736644, -0.05349315,  0.13024579,  0.1666748 ,\n",
              "            0.10085115, -0.22297725, -0.18651742,  0.15428513,\n",
              "           -0.0563795 ,  0.02727103,  0.00113688,  0.18169387],\n",
              "          [-0.05612422, -0.04463722,  0.10158637,  0.02034799,\n",
              "           -0.04356169, -0.2101895 , -0.17018501, -0.22338548,\n",
              "            0.2530376 , -0.06792978,  0.05991254,  0.14828086,\n",
              "           -0.14307164, -0.16819857, -0.03562242,  0.01753301],\n",
              "          [ 0.04963868,  0.03074271, -0.13571097,  0.04124813,\n",
              "            0.01700335, -0.16640438,  0.17027941,  0.06839474,\n",
              "            0.21936238, -0.06213649, -0.06355547, -0.14337613,\n",
              "           -0.01584296,  0.00528579,  0.00537152,  0.10210592]],\n",
              " \n",
              "         [[-0.05140112, -0.15347947, -0.14081039,  0.14350711,\n",
              "            0.11348263, -0.05704068, -0.05141038, -0.11814141,\n",
              "            0.13197526, -0.15579985,  0.07218798, -0.05397184,\n",
              "            0.10321277, -0.15163709, -0.21185453,  0.08340874],\n",
              "          [ 0.00863332, -0.06082493,  0.02451223,  0.11477598,\n",
              "           -0.01391207,  0.01549559,  0.06426721, -0.15842174,\n",
              "            0.14501594, -0.1160922 ,  0.0219058 , -0.04580669,\n",
              "           -0.15765241, -0.00695971, -0.20072833, -0.01137993],\n",
              "          [-0.09048332,  0.0176714 ,  0.09491708, -0.19198021,\n",
              "            0.13065112,  0.114476  ,  0.17840263, -0.15215692,\n",
              "           -0.11557245,  0.03238291, -0.2276151 , -0.16903313,\n",
              "            0.04529225,  0.03407687,  0.0682032 , -0.10355701]],\n",
              " \n",
              "         [[-0.06847084, -0.15718696, -0.08935364, -0.17332995,\n",
              "            0.17800313,  0.26030594, -0.044192  ,  0.1311623 ,\n",
              "           -0.2795981 ,  0.1562169 ,  0.18427148,  0.01678961,\n",
              "           -0.08110964,  0.12918246,  0.01725463, -0.1519887 ],\n",
              "          [ 0.05165625, -0.0190671 , -0.2067906 ,  0.21197367,\n",
              "            0.01090284, -0.06724587, -0.06223952,  0.14574426,\n",
              "           -0.18394075,  0.168713  ,  0.03368711, -0.01248228,\n",
              "           -0.16498642,  0.08012342,  0.23250371, -0.05262294],\n",
              "          [-0.05973453, -0.07663763,  0.08404816, -0.1713072 ,\n",
              "            0.04050018, -0.1591655 ,  0.09894413, -0.04939918,\n",
              "           -0.13664612, -0.01063126,  0.10737205, -0.16626121,\n",
              "            0.05853936, -0.11572332,  0.02995615,  0.09531486]]],\n",
              " \n",
              " \n",
              "        [[[-0.04140931, -0.21341294, -0.08082227, -0.19226591,\n",
              "           -0.08364857, -0.21376266,  0.01580726, -0.10672224,\n",
              "            0.03200867, -0.01274871, -0.06184819, -0.16093448,\n",
              "           -0.1096091 ,  0.06462003,  0.19344229,  0.09271497],\n",
              "          [-0.08837451, -0.02522307,  0.08338313, -0.12090321,\n",
              "           -0.05551589,  0.0747095 , -0.15052289,  0.02134184,\n",
              "            0.23300382, -0.13402009, -0.17946887,  0.15712893,\n",
              "           -0.06701358, -0.20373313, -0.05308695,  0.11407006],\n",
              "          [-0.18957704,  0.05393331, -0.17254472,  0.07661286,\n",
              "           -0.22689882,  0.06516548,  0.08544505,  0.00279893,\n",
              "           -0.14792672, -0.03024858, -0.03143303,  0.11407226,\n",
              "           -0.15279464, -0.05041821, -0.09559733,  0.14580607]],\n",
              " \n",
              "         [[-0.11816262, -0.01184934, -0.09030519, -0.18112871,\n",
              "            0.05099198,  0.12553895,  0.14718436, -0.06992386,\n",
              "           -0.210568  , -0.2241697 , -0.15440634,  0.01697279,\n",
              "           -0.12378131, -0.16561641,  0.02775168, -0.0138314 ],\n",
              "          [ 0.0821247 ,  0.02830783, -0.05978766,  0.20361859,\n",
              "            0.19300634, -0.03695393, -0.02986863, -0.16748573,\n",
              "           -0.22702526, -0.06883721,  0.1295097 ,  0.19081071,\n",
              "            0.00557641, -0.24159119, -0.16281788,  0.1295138 ],\n",
              "          [ 0.14142904, -0.03563416,  0.07737536,  0.05731314,\n",
              "           -0.07488195, -0.00160797,  0.15416765, -0.20986171,\n",
              "            0.03463988,  0.16333622,  0.01716708, -0.11926348,\n",
              "           -0.01168345,  0.1389889 , -0.12398989, -0.22070882]],\n",
              " \n",
              "         [[-0.02512087,  0.10214798, -0.10954835,  0.03830923,\n",
              "            0.10047054,  0.19152926,  0.01886278,  0.22411755,\n",
              "           -0.05378683,  0.14196683,  0.12093055,  0.1463205 ,\n",
              "            0.07151917, -0.1819884 ,  0.14760406, -0.04488141],\n",
              "          [ 0.10153462,  0.19047683,  0.01571959,  0.04095826,\n",
              "            0.06440078,  0.18185578,  0.01974289, -0.09333344,\n",
              "            0.09241575,  0.15480888,  0.22149985,  0.11724184,\n",
              "            0.02244529,  0.10658093,  0.2255021 ,  0.02369262],\n",
              "          [ 0.12882559,  0.21141668, -0.02995334,  0.10767746,\n",
              "           -0.02869586,  0.07096676,  0.05525681,  0.17453253,\n",
              "            0.03878678,  0.06659904,  0.10517017,  0.02863048,\n",
              "            0.17363091, -0.03079828, -0.17518528,  0.10638763]]]],\n",
              "       dtype=float32),\n",
              " array([ 0.00258859,  0.06509493, -0.02976452,  0.06239487, -0.03144287,\n",
              "        -0.09343985, -0.09636946,  0.03628981,  0.03264595,  0.04709627,\n",
              "        -0.02173181,  0.01207149,  0.01290322, -0.00562704,  0.02132414,\n",
              "        -0.09501752], dtype=float32),\n",
              " array([[[[-1.08861933e-02,  5.01905084e-02,  3.91726233e-02, ...,\n",
              "           -4.72420901e-02, -6.44270480e-02,  6.61866069e-02],\n",
              "          [ 2.14705616e-01, -1.21453799e-01,  4.65150960e-02, ...,\n",
              "            5.24737872e-02,  1.01187296e-01, -3.90206277e-02],\n",
              "          [ 1.23807959e-01,  1.52699858e-01,  7.84422606e-02, ...,\n",
              "           -3.08278780e-02,  1.00826271e-01,  3.94534208e-02],\n",
              "          ...,\n",
              "          [-4.79702987e-02,  1.06525242e-01,  9.77504179e-02, ...,\n",
              "           -1.41283348e-01, -4.22988459e-02,  3.04848850e-02],\n",
              "          [-1.75143462e-02, -1.50385812e-01, -2.04456836e-01, ...,\n",
              "            1.57670751e-02,  8.71607438e-02, -4.99072596e-02],\n",
              "          [-2.50654686e-02,  4.56194878e-02, -1.38593808e-01, ...,\n",
              "            9.10142586e-02, -2.58245748e-02,  1.10116757e-01]],\n",
              " \n",
              "         [[-8.33634660e-02, -1.28981322e-01, -5.26511967e-02, ...,\n",
              "            1.96763203e-02,  1.12804761e-02,  3.98153998e-02],\n",
              "          [ 2.01278076e-01, -1.28052793e-02,  2.28380471e-01, ...,\n",
              "           -1.55167595e-01, -1.11559801e-01,  3.83116193e-02],\n",
              "          [ 5.78557178e-02,  3.46005782e-02, -1.02297783e-01, ...,\n",
              "           -1.21406823e-01, -6.11420237e-02, -8.68688449e-02],\n",
              "          ...,\n",
              "          [-1.73287783e-02,  3.32056102e-03,  1.43403206e-02, ...,\n",
              "           -2.93994267e-02, -1.12200022e-01, -3.47522460e-02],\n",
              "          [ 2.75113374e-01, -2.01149732e-02, -2.49901377e-02, ...,\n",
              "            1.68500803e-02,  1.83407560e-01, -3.29541378e-02],\n",
              "          [-1.80009112e-01,  5.32236472e-02,  3.89433987e-02, ...,\n",
              "           -1.43012237e-02,  5.39675243e-02,  1.09268390e-02]],\n",
              " \n",
              "         [[-6.93741739e-02,  8.11677575e-02,  7.76647925e-02, ...,\n",
              "           -6.48846105e-02,  8.07130635e-02,  7.30806589e-02],\n",
              "          [-1.25562266e-01,  2.91999206e-02, -3.81721929e-02, ...,\n",
              "            8.27689692e-02, -9.96485502e-02, -7.49162026e-03],\n",
              "          [-1.21290982e-01, -2.28070952e-02, -1.06865965e-01, ...,\n",
              "            7.26310760e-02,  8.01865682e-02,  1.57341398e-02],\n",
              "          ...,\n",
              "          [ 2.77110282e-02, -8.51276219e-02,  6.38028383e-02, ...,\n",
              "           -1.09681478e-02,  4.41649184e-02,  1.03494033e-01],\n",
              "          [-9.69420746e-02, -2.35005505e-02,  1.44344896e-01, ...,\n",
              "           -9.90990177e-02,  2.49100775e-02,  8.44140947e-02],\n",
              "          [ 1.06063165e-01,  3.25329714e-02,  5.36338203e-02, ...,\n",
              "            3.41435857e-02,  1.76640321e-03, -2.18397826e-02]]],\n",
              " \n",
              " \n",
              "        [[[ 5.93063273e-02,  2.58274414e-02, -7.51267001e-02, ...,\n",
              "           -4.81214561e-02,  5.29050566e-02, -3.00795585e-02],\n",
              "          [ 3.22395653e-01, -1.19983600e-02,  1.04962312e-01, ...,\n",
              "           -1.94760978e-01, -8.69701579e-02,  7.55249560e-02],\n",
              "          [ 8.05099979e-02, -1.53947473e-02, -2.46343296e-02, ...,\n",
              "            5.18731959e-02, -7.65101984e-02, -5.89804538e-02],\n",
              "          ...,\n",
              "          [ 1.80347085e-01,  2.56545339e-02, -5.84281795e-02, ...,\n",
              "            8.10455158e-03, -3.16032581e-02,  1.13895880e-02],\n",
              "          [ 2.41479315e-02, -2.09676828e-02,  5.72355911e-02, ...,\n",
              "            4.99019064e-02,  3.49034905e-04, -3.80210429e-02],\n",
              "          [ 2.72989012e-02, -9.44213346e-02,  2.42659319e-02, ...,\n",
              "            5.93092218e-02, -1.02075264e-01,  8.73334110e-02]],\n",
              " \n",
              "         [[-2.40222327e-02, -1.23856373e-01,  2.07557734e-02, ...,\n",
              "           -6.63490444e-02, -6.15854040e-02, -1.09124690e-01],\n",
              "          [ 2.95446724e-01, -7.73648173e-02,  2.00998083e-01, ...,\n",
              "           -3.76494706e-01, -1.15602963e-01,  3.13922428e-02],\n",
              "          [-1.30793989e-01, -1.30511850e-01, -1.28329232e-01, ...,\n",
              "           -6.45600036e-02, -2.84764525e-02, -3.29848193e-02],\n",
              "          ...,\n",
              "          [-5.64776175e-02, -1.24231741e-01,  9.03908089e-02, ...,\n",
              "           -6.39312044e-02,  4.06861939e-02,  1.22934029e-01],\n",
              "          [ 3.91049266e-01, -2.54613608e-01,  2.72981763e-01, ...,\n",
              "            1.34431571e-01, -1.70853396e-03,  1.20640248e-01],\n",
              "          [-2.72197872e-02,  5.05221821e-02, -1.34978488e-01, ...,\n",
              "            1.11675397e-01,  5.72121777e-02,  8.56293589e-02]],\n",
              " \n",
              "         [[-5.24877347e-02,  4.98851314e-02, -5.32166511e-02, ...,\n",
              "            3.56472805e-02,  1.30771957e-02,  1.05191313e-01],\n",
              "          [ 7.91892484e-02,  1.35233607e-02,  8.56865719e-02, ...,\n",
              "           -3.02817021e-02, -1.73445508e-01, -1.53872117e-01],\n",
              "          [-9.76803228e-02, -7.04920515e-02,  7.22497329e-02, ...,\n",
              "            8.17572400e-02, -1.06609743e-02,  1.71176251e-02],\n",
              "          ...,\n",
              "          [-5.10501117e-02,  5.45682982e-02, -3.28572723e-03, ...,\n",
              "           -2.47875210e-02, -6.88128397e-02,  8.84883627e-02],\n",
              "          [ 1.57911882e-01, -1.01473577e-01,  9.22784433e-02, ...,\n",
              "           -3.45422365e-02,  5.50393984e-02,  1.57180578e-01],\n",
              "          [ 8.55209082e-02,  2.63912864e-02,  2.03413945e-02, ...,\n",
              "            5.07915094e-02,  2.48182267e-02, -6.59720972e-03]]],\n",
              " \n",
              " \n",
              "        [[[-1.51123507e-02,  7.46857002e-02,  1.75975338e-02, ...,\n",
              "            1.05780408e-01, -9.03039798e-02, -6.76898635e-04],\n",
              "          [ 3.79078597e-01,  9.39406604e-02,  1.02157548e-01, ...,\n",
              "           -4.30775613e-01, -3.56436856e-02, -1.25890553e-01],\n",
              "          [-4.60771844e-02,  1.48651293e-02,  1.72986791e-01, ...,\n",
              "           -7.96843693e-02,  1.34439036e-01, -9.42264348e-02],\n",
              "          ...,\n",
              "          [-7.52844242e-03, -6.77963113e-03, -4.54551401e-03, ...,\n",
              "           -1.84601441e-01,  4.02350910e-02, -7.78708607e-02],\n",
              "          [ 1.43623546e-01, -8.39317366e-02, -1.20968662e-01, ...,\n",
              "            4.58308384e-02,  7.80534744e-02,  4.90031429e-02],\n",
              "          [-8.81273150e-02, -1.12422584e-02,  2.78680250e-02, ...,\n",
              "            9.92995426e-02, -1.11482725e-01, -6.85789213e-02]],\n",
              " \n",
              "         [[ 1.17444009e-01,  4.51117977e-02,  3.72261144e-02, ...,\n",
              "            3.60656977e-02,  1.07560955e-01, -1.24799922e-01],\n",
              "          [ 3.38760853e-01,  9.24836174e-02,  1.32505558e-02, ...,\n",
              "           -2.10355625e-01,  4.23132330e-02, -1.26920473e-02],\n",
              "          [ 1.87117071e-03,  1.07656233e-01, -8.63245279e-02, ...,\n",
              "           -6.13444811e-03, -3.14791501e-02, -7.13843331e-02],\n",
              "          ...,\n",
              "          [-1.27525300e-01, -1.20134786e-01, -3.76641639e-02, ...,\n",
              "           -1.33157268e-01, -1.59525163e-02, -8.29113796e-02],\n",
              "          [ 2.49782041e-01, -9.58183408e-03,  2.22192705e-01, ...,\n",
              "           -7.87903089e-03,  1.49216488e-01, -1.05525576e-01],\n",
              "          [-1.40665606e-01, -9.80742052e-02,  5.08788340e-02, ...,\n",
              "            3.96200269e-02,  2.90517043e-02, -1.11779552e-02]],\n",
              " \n",
              "         [[-1.12956479e-01, -5.74793406e-02,  9.66005474e-02, ...,\n",
              "            9.20141265e-02, -8.28010738e-02, -6.88421354e-02],\n",
              "          [ 6.56828377e-03,  1.10356905e-01, -1.73621416e-01, ...,\n",
              "            5.34824990e-02, -1.23860404e-01, -8.58785138e-02],\n",
              "          [ 3.66671905e-02, -1.00507196e-02, -1.19678497e-01, ...,\n",
              "            6.21444061e-02,  5.31832129e-03,  1.28300101e-01],\n",
              "          ...,\n",
              "          [ 1.11329434e-02, -8.23574588e-02, -1.41744837e-01, ...,\n",
              "           -8.31225663e-02, -5.41504994e-02, -5.17390743e-02],\n",
              "          [ 1.60319939e-01, -3.68608907e-02,  2.89389849e-01, ...,\n",
              "            2.07756311e-02,  1.31219178e-01,  9.84977931e-02],\n",
              "          [ 3.75836529e-02, -4.45545353e-02, -9.05101448e-02, ...,\n",
              "           -1.30470777e-02,  1.35962861e-02,  1.04303539e-01]]]],\n",
              "       dtype=float32),\n",
              " array([-0.02750864, -0.04277912, -0.00541971,  0.01032392, -0.03446163,\n",
              "        -0.04258003, -0.00839936, -0.03810481, -0.0484255 , -0.00271349,\n",
              "         0.05164094, -0.06497408, -0.08101346,  0.00283635,  0.0948559 ,\n",
              "        -0.07758386, -0.04092631,  0.03211056,  0.00163169, -0.04517806,\n",
              "         0.00916636, -0.01741   , -0.03273792, -0.04946571, -0.03383309,\n",
              "        -0.03217982,  0.03524986, -0.04688475, -0.06497488, -0.03069756,\n",
              "         0.02312813, -0.02143103], dtype=float32),\n",
              " array([[[[-0.0101611 , -0.00790617,  0.01425773, ..., -0.01632976,\n",
              "            0.0618863 ,  0.00850406],\n",
              "          [-0.03181212,  0.00268689, -0.05305006, ...,  0.04645156,\n",
              "           -0.01595196, -0.04928033],\n",
              "          [-0.04056378, -0.04153441, -0.16145958, ..., -0.07568461,\n",
              "            0.04664865, -0.05285687],\n",
              "          ...,\n",
              "          [ 0.02609055, -0.02311322,  0.14339368, ..., -0.10362966,\n",
              "           -0.12487075, -0.05076878],\n",
              "          [ 0.05161123, -0.12732689,  0.07035807, ..., -0.03824843,\n",
              "            0.03041208,  0.00939673],\n",
              "          [ 0.01769387, -0.03288636, -0.0037378 , ..., -0.02992222,\n",
              "           -0.046219  ,  0.05248574]],\n",
              " \n",
              "         [[-0.06985489,  0.1612906 , -0.06613959, ...,  0.02410812,\n",
              "            0.02836563, -0.06820801],\n",
              "          [-0.01585035, -0.00750984, -0.044556  , ..., -0.03578629,\n",
              "            0.04147393,  0.01951765],\n",
              "          [-0.00062201, -0.0245684 , -0.09424561, ..., -0.06756999,\n",
              "            0.0162602 , -0.13576978],\n",
              "          ...,\n",
              "          [-0.02871645,  0.00594454,  0.01804226, ...,  0.02444379,\n",
              "           -0.10236638, -0.05213202],\n",
              "          [-0.07868824, -0.09441135, -0.06133998, ..., -0.06132799,\n",
              "            0.04953435, -0.07862984],\n",
              "          [ 0.02966126, -0.08545223, -0.06847667, ..., -0.01496073,\n",
              "           -0.00579748, -0.08753542]],\n",
              " \n",
              "         [[-0.07184326,  0.05177965, -0.13428602, ..., -0.07376826,\n",
              "           -0.08661798, -0.01260236],\n",
              "          [ 0.08049252,  0.064035  , -0.06925862, ..., -0.06719054,\n",
              "           -0.0625661 ,  0.07437578],\n",
              "          [-0.08230024, -0.01706891, -0.05230429, ..., -0.01642373,\n",
              "           -0.04932659, -0.05792851],\n",
              "          ...,\n",
              "          [-0.05345825, -0.13022691,  0.00605062, ..., -0.04732386,\n",
              "           -0.00507565, -0.01814679],\n",
              "          [-0.09136915, -0.16698632,  0.08272402, ..., -0.05002675,\n",
              "            0.01215095, -0.04676621],\n",
              "          [ 0.01571622, -0.09625022, -0.036467  , ...,  0.01448012,\n",
              "           -0.04926015,  0.01655926]]],\n",
              " \n",
              " \n",
              "        [[[-0.07792193, -0.06794672,  0.01576366, ..., -0.02058253,\n",
              "           -0.06798491,  0.0669959 ],\n",
              "          [-0.03259603, -0.02954322,  0.07832836, ...,  0.00585171,\n",
              "            0.03355305, -0.03125996],\n",
              "          [ 0.05716958, -0.00956428, -0.15850349, ..., -0.13340439,\n",
              "            0.05435979, -0.00529512],\n",
              "          ...,\n",
              "          [ 0.00242772,  0.03249256,  0.00930732, ..., -0.12036268,\n",
              "           -0.1401269 , -0.1016159 ],\n",
              "          [ 0.03972867, -0.1093836 , -0.04874247, ...,  0.0311476 ,\n",
              "           -0.07541067, -0.08967756],\n",
              "          [ 0.05778255,  0.01867857,  0.09777907, ..., -0.09992472,\n",
              "           -0.09149106,  0.05870859]],\n",
              " \n",
              "         [[ 0.02239633,  0.07085916,  0.01296374, ..., -0.05298508,\n",
              "            0.16734228, -0.0254469 ],\n",
              "          [-0.0292996 , -0.07194453, -0.03931879, ...,  0.05158715,\n",
              "           -0.07051931,  0.03210852],\n",
              "          [-0.09339719,  0.12012324,  0.16191775, ..., -0.0612434 ,\n",
              "            0.0627573 , -0.12294288],\n",
              "          ...,\n",
              "          [ 0.07410653, -0.07043925, -0.11939564, ..., -0.13627794,\n",
              "           -0.15087809,  0.05571987],\n",
              "          [ 0.04377514, -0.08215431,  0.04768874, ..., -0.08022013,\n",
              "            0.11245719, -0.06145179],\n",
              "          [ 0.07374346, -0.0159107 , -0.01953479, ...,  0.00454247,\n",
              "           -0.02259045, -0.01860926]],\n",
              " \n",
              "         [[-0.0480313 ,  0.1121779 , -0.10275597, ...,  0.04825199,\n",
              "           -0.02945866,  0.09230656],\n",
              "          [ 0.05161139, -0.04376761, -0.03471683, ...,  0.04941719,\n",
              "            0.04130558,  0.00501298],\n",
              "          [-0.00421502, -0.03129062,  0.1034739 , ..., -0.08259697,\n",
              "           -0.01997313, -0.01450586],\n",
              "          ...,\n",
              "          [-0.06716612,  0.06279835, -0.06807525, ..., -0.00373629,\n",
              "            0.06497996, -0.07217446],\n",
              "          [-0.02979358,  0.04400397, -0.05461818, ...,  0.0420789 ,\n",
              "           -0.02992778, -0.0917954 ],\n",
              "          [ 0.08503943, -0.06160726,  0.01444956, ..., -0.0019655 ,\n",
              "           -0.03828383, -0.03040553]]],\n",
              " \n",
              " \n",
              "        [[[ 0.11213703, -0.18847564,  0.03180786, ...,  0.03405731,\n",
              "           -0.02057007,  0.01796424],\n",
              "          [-0.04325666,  0.06679276, -0.0281381 , ...,  0.06120803,\n",
              "            0.03897887,  0.06813193],\n",
              "          [-0.03216149, -0.05881945,  0.00172609, ..., -0.16920018,\n",
              "           -0.02542829,  0.03283879],\n",
              "          ...,\n",
              "          [-0.04971543, -0.02729169,  0.0444594 , ...,  0.00389245,\n",
              "            0.00211606, -0.06470253],\n",
              "          [-0.05024416, -0.04742814, -0.08041163, ..., -0.06834294,\n",
              "            0.05405279,  0.00552206],\n",
              "          [ 0.04467026,  0.10685993,  0.19509651, ..., -0.10502072,\n",
              "            0.02289118, -0.03022775]],\n",
              " \n",
              "         [[ 0.0364088 ,  0.04865477,  0.08508933, ..., -0.03640368,\n",
              "            0.18506524,  0.03952534],\n",
              "          [-0.06391846,  0.05981423,  0.00194475, ..., -0.02032534,\n",
              "           -0.05720446, -0.05626142],\n",
              "          [-0.05690103, -0.03526624,  0.17391072, ..., -0.07684221,\n",
              "            0.00851837, -0.10613188],\n",
              "          ...,\n",
              "          [ 0.08855679,  0.02485191, -0.0124695 , ..., -0.01177009,\n",
              "           -0.16061182,  0.03746136],\n",
              "          [-0.05571322,  0.00388543, -0.01150156, ..., -0.03543817,\n",
              "            0.12161416, -0.01485737],\n",
              "          [ 0.11057636, -0.01571956,  0.03096625, ..., -0.08852106,\n",
              "           -0.04931316,  0.02088538]],\n",
              " \n",
              "         [[-0.1426668 ,  0.09040464,  0.06284785, ...,  0.08108792,\n",
              "            0.02180802,  0.07168081],\n",
              "          [-0.05893487, -0.06803069, -0.02113087, ..., -0.07010096,\n",
              "            0.06609746,  0.0276428 ],\n",
              "          [-0.09116096, -0.0808012 ,  0.13532424, ..., -0.10048613,\n",
              "            0.07113   , -0.06041759],\n",
              "          ...,\n",
              "          [-0.06674518,  0.10478451, -0.02904509, ..., -0.06798577,\n",
              "           -0.03373638, -0.01075416],\n",
              "          [-0.02379115, -0.04940556, -0.09351628, ...,  0.02080672,\n",
              "            0.02671869, -0.04052867],\n",
              "          [-0.07169119, -0.06885587, -0.08100428, ..., -0.00029513,\n",
              "            0.05782438, -0.07431617]]]], dtype=float32),\n",
              " array([-0.04736576, -0.06292881, -0.04002281, -0.02637298, -0.0519745 ,\n",
              "        -0.07499909, -0.0634216 , -0.03324316, -0.13586734, -0.07323259,\n",
              "        -0.11490995, -0.09398489, -0.01479972, -0.07000666, -0.03715025,\n",
              "        -0.06730875, -0.05034104, -0.07094351,  0.00189929, -0.01391762,\n",
              "        -0.01080511, -0.00767684, -0.08332593, -0.08462493, -0.0289533 ,\n",
              "        -0.11921872, -0.03799381, -0.03628815, -0.03859737, -0.03341347,\n",
              "        -0.04057802, -0.01326786, -0.02575525,  0.00601229, -0.0470731 ,\n",
              "        -0.04175366,  0.01632898, -0.02998277, -0.06369378, -0.04817748,\n",
              "         0.01630259, -0.06657393, -0.02142508, -0.049956  , -0.07779945,\n",
              "        -0.05049285, -0.01546685, -0.0764537 , -0.03206791,  0.01940887,\n",
              "        -0.10344668,  0.00126449, -0.05992368, -0.01204756, -0.06462685,\n",
              "         0.05562161, -0.03718335, -0.02453628, -0.06607135, -0.01828592,\n",
              "        -0.01090008, -0.02961851, -0.04776448, -0.02519721], dtype=float32),\n",
              " array([[-0.02156935, -0.00485899, -0.04593839, ..., -0.06261214,\n",
              "          0.00795087,  0.08138146],\n",
              "        [-0.01506856, -0.00155434,  0.07280833, ...,  0.09278908,\n",
              "         -0.0169415 , -0.08264825],\n",
              "        [ 0.00069374,  0.01525015,  0.11058678, ...,  0.10416345,\n",
              "         -0.01591315, -0.04578106],\n",
              "        ...,\n",
              "        [-0.01978668, -0.0156214 , -0.03808788, ..., -0.03886779,\n",
              "         -0.01837786, -0.035322  ],\n",
              "        [ 0.0018261 , -0.00096613,  0.00711191, ..., -0.0107705 ,\n",
              "         -0.01557264, -0.01552141],\n",
              "        [ 0.00679367,  0.00637791, -0.04558755, ..., -0.02206814,\n",
              "          0.00328635,  0.02622206]], dtype=float32),\n",
              " array([-0.00531755, -0.00599704,  0.08498215, -0.0050008 ,  0.01300661,\n",
              "        -0.0059887 , -0.00544238, -0.0185938 , -0.00555761, -0.03303601,\n",
              "        -0.00599895, -0.00577662, -0.00595586, -0.00600395, -0.00548664,\n",
              "         0.04798001, -0.01516181, -0.0135141 ,  0.07651733, -0.00538439,\n",
              "        -0.00598284,  0.07600632,  0.08551406,  0.08233579,  0.00134333,\n",
              "         0.07240248, -0.01833433, -0.00681961, -0.01884769, -0.01859666,\n",
              "        -0.00600313, -0.00489891, -0.00888455, -0.00539327, -0.04494555,\n",
              "        -0.00627277, -0.0137452 ,  0.06977075, -0.00690804, -0.01321745,\n",
              "        -0.01783131, -0.00599146, -0.00600303,  0.06637561, -0.02247556,\n",
              "        -0.00600311, -0.00600077,  0.07075457, -0.00530862, -0.00543901,\n",
              "        -0.00600303, -0.00600056,  0.02547102, -0.00600123, -0.03498944,\n",
              "        -0.00898713, -0.00535207, -0.01064654, -0.00636957,  0.07171211,\n",
              "        -0.02133913, -0.01409468, -0.00600306, -0.00488017,  0.12113252,\n",
              "        -0.04108341, -0.0311157 , -0.0051117 , -0.00599799, -0.01546088,\n",
              "        -0.03982201, -0.00600082,  0.04886666, -0.00698104, -0.01057898,\n",
              "        -0.01845311, -0.04451795, -0.00533544, -0.00540633, -0.00477987,\n",
              "        -0.00599217, -0.03896212,  0.00166449, -0.00602972, -0.020303  ,\n",
              "        -0.00625909,  0.06836107, -0.00883701, -0.00600262, -0.00146223,\n",
              "        -0.01976459,  0.03362743, -0.00938158, -0.00599523, -0.00832265,\n",
              "        -0.02944335, -0.00538518,  0.06770662,  0.01707507, -0.01180253,\n",
              "        -0.00600059, -0.00600321,  0.07997946,  0.04393948, -0.02848158,\n",
              "        -0.00600016, -0.00542608, -0.00555775, -0.00572457,  0.07541155,\n",
              "         0.01441819, -0.00600255,  0.06856836, -0.00665491,  0.01201624,\n",
              "         0.02649884, -0.0060044 , -0.00555825, -0.00600455,  0.0786116 ,\n",
              "        -0.01368086, -0.0580855 , -0.00711966, -0.00609322,  0.0780021 ,\n",
              "         0.06099765, -0.00590793, -0.00485874], dtype=float32),\n",
              " array([[ 0.00492957],\n",
              "        [-0.04975232],\n",
              "        [ 0.28670654],\n",
              "        [ 0.09314483],\n",
              "        [-0.22380506],\n",
              "        [-0.02062762],\n",
              "        [ 0.14887232],\n",
              "        [-0.03572864],\n",
              "        [ 0.05906649],\n",
              "        [-0.09116435],\n",
              "        [-0.06587969],\n",
              "        [-0.01774199],\n",
              "        [-0.0034817 ],\n",
              "        [-0.16884112],\n",
              "        [ 0.01273971],\n",
              "        [-0.30802345],\n",
              "        [-0.01408489],\n",
              "        [-0.30469403],\n",
              "        [ 0.23156561],\n",
              "        [ 0.07507544],\n",
              "        [-0.01484134],\n",
              "        [ 0.26160765],\n",
              "        [ 0.22583604],\n",
              "        [ 0.23944417],\n",
              "        [-0.01213566],\n",
              "        [ 0.261713  ],\n",
              "        [-0.17065752],\n",
              "        [-0.00421719],\n",
              "        [-0.1717409 ],\n",
              "        [-0.00527131],\n",
              "        [-0.10931146],\n",
              "        [ 0.17406091],\n",
              "        [-0.23520696],\n",
              "        [ 0.10169606],\n",
              "        [ 0.09041532],\n",
              "        [ 0.12326419],\n",
              "        [-0.01819234],\n",
              "        [ 0.10216664],\n",
              "        [-0.09038762],\n",
              "        [ 0.10249484],\n",
              "        [-0.07287337],\n",
              "        [-0.04894316],\n",
              "        [-0.16838382],\n",
              "        [-0.26800427],\n",
              "        [ 0.26377645],\n",
              "        [-0.18338951],\n",
              "        [-0.09333009],\n",
              "        [ 0.13220693],\n",
              "        [ 0.00329148],\n",
              "        [ 0.11876109],\n",
              "        [-0.18230121],\n",
              "        [-0.08918856],\n",
              "        [-0.316679  ],\n",
              "        [-0.11580326],\n",
              "        [ 0.10718121],\n",
              "        [-0.04637032],\n",
              "        [ 0.05054475],\n",
              "        [ 0.01421953],\n",
              "        [-0.16208887],\n",
              "        [ 0.14204498],\n",
              "        [-0.19829905],\n",
              "        [ 0.17898323],\n",
              "        [-0.18485317],\n",
              "        [ 0.14915326],\n",
              "        [ 0.17434575],\n",
              "        [ 0.09596608],\n",
              "        [ 0.02925464],\n",
              "        [-0.19408026],\n",
              "        [-0.05670279],\n",
              "        [-0.16413787],\n",
              "        [-0.03147538],\n",
              "        [-0.15418199],\n",
              "        [ 0.22779997],\n",
              "        [-0.15228699],\n",
              "        [-0.28037107],\n",
              "        [-0.1413661 ],\n",
              "        [ 0.00416995],\n",
              "        [ 0.04787042],\n",
              "        [ 0.05076848],\n",
              "        [-0.00426025],\n",
              "        [-0.05198392],\n",
              "        [-0.05354165],\n",
              "        [ 0.0195368 ],\n",
              "        [-0.08608753],\n",
              "        [-0.19502427],\n",
              "        [ 0.05780197],\n",
              "        [ 0.08814216],\n",
              "        [-0.12361188],\n",
              "        [-0.15598902],\n",
              "        [ 0.0992432 ],\n",
              "        [-0.00249233],\n",
              "        [-0.29436943],\n",
              "        [-0.10140224],\n",
              "        [-0.03991506],\n",
              "        [-0.24171935],\n",
              "        [-0.21484235],\n",
              "        [ 0.0773802 ],\n",
              "        [ 0.1403242 ],\n",
              "        [-0.26143372],\n",
              "        [ 0.05865375],\n",
              "        [ 0.0684398 ],\n",
              "        [-0.19702514],\n",
              "        [ 0.12058897],\n",
              "        [ 0.18822077],\n",
              "        [ 0.11681984],\n",
              "        [-0.07148138],\n",
              "        [ 0.1004763 ],\n",
              "        [ 0.07763375],\n",
              "        [ 0.00646228],\n",
              "        [ 0.23923293],\n",
              "        [ 0.01330806],\n",
              "        [-0.15201497],\n",
              "        [ 0.16941571],\n",
              "        [ 0.12094392],\n",
              "        [-0.3078733 ],\n",
              "        [ 0.07056218],\n",
              "        [-0.1605563 ],\n",
              "        [ 0.04965607],\n",
              "        [-0.20304191],\n",
              "        [ 0.16090317],\n",
              "        [ 0.02997012],\n",
              "        [-0.05412311],\n",
              "        [ 0.05756404],\n",
              "        [-0.00834541],\n",
              "        [ 0.15923873],\n",
              "        [ 0.29605904],\n",
              "        [-0.0107821 ],\n",
              "        [-0.22729178]], dtype=float32),\n",
              " array([0.00935132], dtype=float32)]"
            ]
          },
          "metadata": {},
          "execution_count": 48
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HuxjW35RFCZ6"
      },
      "source": [
        "<h3>Evaluasi Model</h3>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rN710zUjRAA8"
      },
      "source": [
        "%matplotlib inline\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "from keras.models import load_model\n",
        "from sklearn.metrics import classification_report, confusion_matrix"
      ],
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 316
        },
        "id": "1AM3ms3zE_AU",
        "outputId": "59046892-2e75-452a-9fff-2cd99459a9ba"
      },
      "source": [
        "plt.style.use(\"ggplot\")\n",
        "plt.figure()\n",
        "plt.plot(np.arange(0, 25), H.history[\"loss\"], label=\"train_loss\")\n",
        "plt.plot(np.arange(0, 25), H.history[\"val_loss\"], label=\"val_loss\")\n",
        "plt.title(\"Loss Plot\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()\n",
        "plt.show()\n",
        "\n",
        "plt.savefig('loss-plot-cancer.png')"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 316
        },
        "id": "q1SJWudcFEDL",
        "outputId": "637ea1bb-e763-422d-c108-8e25c1b6bc66"
      },
      "source": [
        "plt.style.use(\"ggplot\")\n",
        "plt.figure()\n",
        "plt.plot(np.arange(0, 25), H.history[\"acc\"], label=\"train_acc\")\n",
        "plt.plot(np.arange(0, 25), H.history[\"val_acc\"], label=\"val_acc\")\n",
        "plt.title(\"Accuracy Plot\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Acc\")\n",
        "plt.legend()\n",
        "plt.show()\n",
        "\n",
        "plt.savefig('accuracy-plot-cancer.png')"
      ],
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H_8mB3igFG-x",
        "outputId": "c4853923-bdd0-4f0d-c6ee-87d3895b598c"
      },
      "source": [
        "from sklearn.metrics import classification_report\n",
        "\n",
        "pred = model.predict(x_val)\n",
        "labels = (pred > 0.5).astype(np.int)\n",
        "\n",
        "print(classification_report(y_val, labels))"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.84      0.84      0.84       360\n",
            "           1       0.81      0.81      0.81       300\n",
            "\n",
            "    accuracy                           0.83       660\n",
            "   macro avg       0.82      0.82      0.82       660\n",
            "weighted avg       0.83      0.83      0.83       660\n",
            "\n"
          ]
        }
      ]
    }
  ]
}