KerasCNNcifar10.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "KerasCNNcifar10.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "EKgMJV1BNQ3D",
        "colab_type": "code",
        "outputId": "1b00c556-130a-4e94-9a78-0d43d3b97107",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "from __future__ import print_function\n",
        "import keras\n",
        "import numpy as np\n",
        "from keras import optimizers\n",
        "from keras.datasets import cifar10\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout, Activation, Flatten\n",
        "from keras.layers import Conv2D, MaxPooling2D\n",
        "import os"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-fBDBjB4PUn2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "batch_size = 256\n",
        "num_classes = 10\n",
        "epochs = 55\n",
        "data_augmentation = True\n",
        "num_predictions = 20\n",
        "save_dir = os.path.join(os.getcwd(), 'saved_models')\n",
        "model_name = 'keras_cifar10_trained_model.h5'"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kvGYKPr6Pxnz",
        "colab_type": "code",
        "outputId": "399e329c-8501-4a89-9631-4b2f3d2eb1a9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        }
      },
      "source": [
        "# The data, split between train and test sets:\n",
        "(x_train, y_train), (x_test, y_test) = cifar10.load_data()\n",
        "print('x_train shape:', x_train.shape)\n",
        "print(x_train.shape[0], 'train samples')\n",
        "print(x_test.shape[0], 'test samples')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n",
            "170500096/170498071 [==============================] - 3s 0us/step\n",
            "x_train shape: (50000, 32, 32, 3)\n",
            "50000 train samples\n",
            "10000 test samples\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "URqMZmezTOrw",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Convert class vectors to binary class matrices.\n",
        "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
        "y_test = keras.utils.to_categorical(y_test, num_classes)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yG9AfXjHcNJT",
        "colab_type": "code",
        "outputId": "8b59d738-fa1e-455a-a35f-8beda057dffe",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 938
        }
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Conv2D(32, (3, 3),\n",
        "                 input_shape=x_train.shape[1:]))\n",
        "model.add(Activation('relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Conv2D(64, (3, 3)))\n",
        "model.add(Activation('relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Dropout(0.25))\n",
        "\n",
        "model.add(Conv2D(128, (3, 3)))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(0.25))\n",
        "\n",
        "model.add(Flatten())\n",
        "model.add(Dense(512))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(0.5))\n",
        "model.add(Dense(num_classes))\n",
        "model.add(Activation('softmax'))\n",
        "\n",
        "model.summary()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:66: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:541: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:4432: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:4267: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:148: The name tf.placeholder_with_default is deprecated. Please use tf.compat.v1.placeholder_with_default instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3733: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n",
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_1 (Conv2D)            (None, 30, 30, 32)        896       \n",
            "_________________________________________________________________\n",
            "activation_1 (Activation)    (None, 30, 30, 32)        0         \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 15, 15, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 13, 13, 64)        18496     \n",
            "_________________________________________________________________\n",
            "activation_2 (Activation)    (None, 13, 13, 64)        0         \n",
            "_________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2 (None, 6, 6, 64)          0         \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 6, 6, 64)          0         \n",
            "_________________________________________________________________\n",
            "conv2d_3 (Conv2D)            (None, 4, 4, 128)         73856     \n",
            "_________________________________________________________________\n",
            "activation_3 (Activation)    (None, 4, 4, 128)         0         \n",
            "_________________________________________________________________\n",
            "dropout_2 (Dropout)          (None, 4, 4, 128)         0         \n",
            "_________________________________________________________________\n",
            "flatten_1 (Flatten)          (None, 2048)              0         \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 512)               1049088   \n",
            "_________________________________________________________________\n",
            "activation_4 (Activation)    (None, 512)               0         \n",
            "_________________________________________________________________\n",
            "dropout_3 (Dropout)          (None, 512)               0         \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 10)                5130      \n",
            "_________________________________________________________________\n",
            "activation_5 (Activation)    (None, 10)                0         \n",
            "=================================================================\n",
            "Total params: 1,147,466\n",
            "Trainable params: 1,147,466\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KAesFdeXRLie",
        "colab_type": "code",
        "outputId": "45b318ae-1827-4b0c-9812-a9542c6c1021",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 122
        }
      },
      "source": [
        "# initiate optimizer\n",
        "opt = keras.optimizers.Adam(lr=0.001,beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.0, amsgrad=False)\n",
        "\n",
        "# Let's train the model using optimiser\n",
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer=opt,\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "x_train = x_train.astype('float32')\n",
        "x_test = x_test.astype('float32')\n",
        "x_train /= 255\n",
        "x_test /= 255\n",
        "\n",
        "print(\"X_train shape:\", x_train.shape)\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/optimizers.py:793: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3576: The name tf.log is deprecated. Please use tf.math.log instead.\n",
            "\n",
            "X_train shape: (50000, 32, 32, 3)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZkTnx01oST40",
        "colab_type": "code",
        "outputId": "acec5c00-72d7-47f9-e5a5-ad7af56b5381",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "if not data_augmentation:\n",
        "    print('Not using data augmentation.')\n",
        "    model.fit(x_train, y_train,\n",
        "              batch_size=batch_size,\n",
        "              epochs=epochs,\n",
        "              validation_data=(x_test, y_test),\n",
        "              shuffle=True)\n",
        "else:\n",
        "    print('Using real-time data augmentation.')\n",
        "    datagen = ImageDataGenerator(\n",
        "        featurewise_center=False, \n",
        "        samplewise_center=False,  \n",
        "        featurewise_std_normalization=False, \n",
        "        samplewise_std_normalization=False, \n",
        "        zca_whitening=False,  \n",
        "        zca_epsilon=1e-06,  \n",
        "        rotation_range=0, \n",
        "       \n",
        "        width_shift_range=0.1,\n",
        "        height_shift_range=0.1,\n",
        "        shear_range=0., \n",
        "        zoom_range=0.,  \n",
        "        channel_shift_range=0.,  \n",
        "        \n",
        "        fill_mode='nearest',\n",
        "        cval=0.,\n",
        "        horizontal_flip=True, \n",
        "        vertical_flip=False,  \n",
        "        \n",
        "        rescale=None,        \n",
        "        preprocessing_function=None,\n",
        "        data_format=None,\n",
        "        validation_split=0.0)\n",
        "\n",
        "    datagen.fit(x_train)\n",
        "\n",
        "    # Fit the model on the batches generated by datagen.flow().\n",
        "    history = model.fit_generator(datagen.flow(x_train, y_train,\n",
        "                                     batch_size=batch_size),\n",
        "                        epochs=epochs,\n",
        "                        validation_data=(x_test, y_test),\n",
        "                        workers=4)\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using real-time data augmentation.\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/math_grad.py:1250: add_dispatch_support.<locals>.wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
            "Epoch 1/55\n",
            "196/196 [==============================] - 40s 206ms/step - loss: 1.7688 - acc: 0.3415 - val_loss: 1.4430 - val_acc: 0.4825\n",
            "Epoch 2/55\n",
            "196/196 [==============================] - 34s 174ms/step - loss: 1.4691 - acc: 0.4658 - val_loss: 1.2353 - val_acc: 0.5587\n",
            "Epoch 3/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 1.3432 - acc: 0.5184 - val_loss: 1.1262 - val_acc: 0.6012\n",
            "Epoch 4/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 1.2534 - acc: 0.5518 - val_loss: 1.0656 - val_acc: 0.6147\n",
            "Epoch 5/55\n",
            "196/196 [==============================] - 34s 174ms/step - loss: 1.1786 - acc: 0.5790 - val_loss: 1.0698 - val_acc: 0.6201\n",
            "Epoch 6/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 1.1313 - acc: 0.6019 - val_loss: 0.9962 - val_acc: 0.6481\n",
            "Epoch 7/55\n",
            "196/196 [==============================] - 34s 174ms/step - loss: 1.0869 - acc: 0.6175 - val_loss: 0.9105 - val_acc: 0.6792\n",
            "Epoch 8/55\n",
            "196/196 [==============================] - 34s 174ms/step - loss: 1.0499 - acc: 0.6293 - val_loss: 0.9050 - val_acc: 0.6815\n",
            "Epoch 9/55\n",
            "196/196 [==============================] - 34s 174ms/step - loss: 1.0180 - acc: 0.6418 - val_loss: 0.8635 - val_acc: 0.6977\n",
            "Epoch 10/55\n",
            "196/196 [==============================] - 34s 174ms/step - loss: 0.9890 - acc: 0.6540 - val_loss: 0.8932 - val_acc: 0.6885\n",
            "Epoch 11/55\n",
            "196/196 [==============================] - 34s 174ms/step - loss: 0.9723 - acc: 0.6600 - val_loss: 0.8218 - val_acc: 0.7127\n",
            "Epoch 12/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 0.9465 - acc: 0.6701 - val_loss: 0.8013 - val_acc: 0.7244\n",
            "Epoch 13/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 0.9277 - acc: 0.6726 - val_loss: 0.7648 - val_acc: 0.7355\n",
            "Epoch 14/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 0.9175 - acc: 0.6775 - val_loss: 0.7906 - val_acc: 0.7221\n",
            "Epoch 15/55\n",
            "196/196 [==============================] - 34s 175ms/step - loss: 0.8935 - acc: 0.6870 - val_loss: 0.7645 - val_acc: 0.7299\n",
            "Epoch 16/55\n",
            "196/196 [==============================] - 34s 171ms/step - loss: 0.8850 - acc: 0.6883 - val_loss: 0.7343 - val_acc: 0.7456\n",
            "Epoch 17/55\n",
            "196/196 [==============================] - 33s 169ms/step - loss: 0.8664 - acc: 0.6975 - val_loss: 0.7714 - val_acc: 0.7278\n",
            "Epoch 18/55\n",
            "196/196 [==============================] - 34s 175ms/step - loss: 0.8587 - acc: 0.6996 - val_loss: 0.6912 - val_acc: 0.7565\n",
            "Epoch 19/55\n",
            "196/196 [==============================] - 34s 171ms/step - loss: 0.8444 - acc: 0.7068 - val_loss: 0.7068 - val_acc: 0.7538\n",
            "Epoch 20/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.8406 - acc: 0.7065 - val_loss: 0.7164 - val_acc: 0.7524\n",
            "Epoch 21/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.8258 - acc: 0.7121 - val_loss: 0.7073 - val_acc: 0.7539\n",
            "Epoch 22/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.8122 - acc: 0.7149 - val_loss: 0.7054 - val_acc: 0.7557\n",
            "Epoch 23/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.8081 - acc: 0.7179 - val_loss: 0.6638 - val_acc: 0.7683\n",
            "Epoch 24/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 0.7994 - acc: 0.7229 - val_loss: 0.6555 - val_acc: 0.7699\n",
            "Epoch 25/55\n",
            "196/196 [==============================] - 33s 170ms/step - loss: 0.7912 - acc: 0.7259 - val_loss: 0.6684 - val_acc: 0.7656\n",
            "Epoch 26/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.7749 - acc: 0.7304 - val_loss: 0.6631 - val_acc: 0.7691\n",
            "Epoch 27/55\n",
            "196/196 [==============================] - 35s 179ms/step - loss: 0.7723 - acc: 0.7307 - val_loss: 0.6836 - val_acc: 0.7600\n",
            "Epoch 28/55\n",
            "196/196 [==============================] - 36s 183ms/step - loss: 0.7697 - acc: 0.7329 - val_loss: 0.6453 - val_acc: 0.7787\n",
            "Epoch 29/55\n",
            "196/196 [==============================] - 35s 178ms/step - loss: 0.7580 - acc: 0.7362 - val_loss: 0.6788 - val_acc: 0.7653\n",
            "Epoch 30/55\n",
            "196/196 [==============================] - 34s 171ms/step - loss: 0.7559 - acc: 0.7351 - val_loss: 0.6577 - val_acc: 0.7706\n",
            "Epoch 31/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.7528 - acc: 0.7387 - val_loss: 0.6217 - val_acc: 0.7797\n",
            "Epoch 32/55\n",
            "196/196 [==============================] - 33s 170ms/step - loss: 0.7427 - acc: 0.7411 - val_loss: 0.6147 - val_acc: 0.7842\n",
            "Epoch 33/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 0.7350 - acc: 0.7428 - val_loss: 0.6197 - val_acc: 0.7844\n",
            "Epoch 34/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.7386 - acc: 0.7425 - val_loss: 0.6621 - val_acc: 0.7730\n",
            "Epoch 35/55\n",
            "196/196 [==============================] - 33s 170ms/step - loss: 0.7217 - acc: 0.7473 - val_loss: 0.6066 - val_acc: 0.7875\n",
            "Epoch 36/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.7255 - acc: 0.7458 - val_loss: 0.6291 - val_acc: 0.7827\n",
            "Epoch 37/55\n",
            "196/196 [==============================] - 33s 170ms/step - loss: 0.7186 - acc: 0.7503 - val_loss: 0.6036 - val_acc: 0.7903\n",
            "Epoch 38/55\n",
            "196/196 [==============================] - 34s 171ms/step - loss: 0.7132 - acc: 0.7507 - val_loss: 0.5904 - val_acc: 0.7970\n",
            "Epoch 39/55\n",
            "196/196 [==============================] - 33s 170ms/step - loss: 0.7162 - acc: 0.7504 - val_loss: 0.6177 - val_acc: 0.7877\n",
            "Epoch 40/55\n",
            "196/196 [==============================] - 33s 170ms/step - loss: 0.7132 - acc: 0.7508 - val_loss: 0.6284 - val_acc: 0.7780\n",
            "Epoch 41/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.7017 - acc: 0.7531 - val_loss: 0.6056 - val_acc: 0.7926\n",
            "Epoch 42/55\n",
            "196/196 [==============================] - 34s 174ms/step - loss: 0.6947 - acc: 0.7551 - val_loss: 0.5856 - val_acc: 0.7916\n",
            "Epoch 43/55\n",
            "196/196 [==============================] - 34s 174ms/step - loss: 0.6995 - acc: 0.7538 - val_loss: 0.6170 - val_acc: 0.7874\n",
            "Epoch 44/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.6935 - acc: 0.7574 - val_loss: 0.6039 - val_acc: 0.7928\n",
            "Epoch 45/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 0.6782 - acc: 0.7633 - val_loss: 0.5917 - val_acc: 0.7951\n",
            "Epoch 46/55\n",
            "196/196 [==============================] - 34s 171ms/step - loss: 0.6880 - acc: 0.7595 - val_loss: 0.5777 - val_acc: 0.8003\n",
            "Epoch 47/55\n",
            "196/196 [==============================] - 33s 170ms/step - loss: 0.6807 - acc: 0.7619 - val_loss: 0.5879 - val_acc: 0.7947\n",
            "Epoch 48/55\n",
            "196/196 [==============================] - 33s 171ms/step - loss: 0.6747 - acc: 0.7640 - val_loss: 0.5949 - val_acc: 0.7894\n",
            "Epoch 49/55\n",
            "196/196 [==============================] - 34s 171ms/step - loss: 0.6716 - acc: 0.7651 - val_loss: 0.5870 - val_acc: 0.7953\n",
            "Epoch 50/55\n",
            "196/196 [==============================] - 33s 171ms/step - loss: 0.6659 - acc: 0.7670 - val_loss: 0.5678 - val_acc: 0.8031\n",
            "Epoch 51/55\n",
            "196/196 [==============================] - 34s 172ms/step - loss: 0.6589 - acc: 0.7687 - val_loss: 0.5767 - val_acc: 0.8017\n",
            "Epoch 52/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 0.6630 - acc: 0.7685 - val_loss: 0.5793 - val_acc: 0.7982\n",
            "Epoch 53/55\n",
            "196/196 [==============================] - 34s 171ms/step - loss: 0.6594 - acc: 0.7697 - val_loss: 0.5702 - val_acc: 0.8049\n",
            "Epoch 54/55\n",
            "196/196 [==============================] - 34s 173ms/step - loss: 0.6585 - acc: 0.7706 - val_loss: 0.6020 - val_acc: 0.7913\n",
            "Epoch 55/55\n",
            "196/196 [==============================] - 33s 171ms/step - loss: 0.6537 - acc: 0.7718 - val_loss: 0.5803 - val_acc: 0.7958\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QpmcTLRSR_KI",
        "colab_type": "code",
        "outputId": "046e4fee-1b92-40b4-e704-6fc9eafd2a81",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        }
      },
      "source": [
        "# Save model and weights\n",
        "if not os.path.isdir(save_dir):\n",
        "    os.makedirs(save_dir)\n",
        "model_path = os.path.join(save_dir, model_name)\n",
        "model.save(model_path)\n",
        "print('Saved trained model at %s ' % model_path)\n",
        "\n",
        "# Score trained model.\n",
        "scores = model.evaluate(x_test, y_test, verbose=1)\n",
        "print('Test loss:', scores[0])\n",
        "print('Test accuracy:', scores[1])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Saved trained model at /content/saved_models/keras_cifar10_trained_model.h5 \n",
            "10000/10000 [==============================] - 1s 122us/step\n",
            "Test loss: 0.5803181800365448\n",
            "Test accuracy: 0.7958\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "W9L2NemRZ1eV",
        "colab_type": "code",
        "outputId": "cb8fc922-b830-432a-948e-b7384a9246d2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 590
        }
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "# list all data in history\n",
        "print(history.history.keys())\n",
        "# summarize history for accuracy\n",
        "plt.plot(history.history['acc'])\n",
        "plt.plot(history.history['val_acc'])\n",
        "plt.title('model accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'test'], loc='upper left')\n",
        "plt.show()\n",
        "# summarize history for loss\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'test'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "dict_keys(['val_loss', 'val_acc', 'loss', 'acc'])\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nlFJK1dGkoJRSqs7/B8hgsC7HmcQsAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}