CatsAndDogs_SmallDataset_Augmentation.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "CatsAndDogs_SmallDataset_Augmentation.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyN6E/pLBWYtpzH7cu1FADrx"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "lY80Tps9d0qW",
        "colab_type": "code",
        "outputId": "62073339-ef3d-4fcc-d1f9-d5aa5fd8cfd7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "import tensorflow as tf\n",
        "import keras"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7NJA-lLt-uen",
        "colab_type": "code",
        "outputId": "103250d0-0e4f-4aab-963d-9703079863b0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        }
      },
      "source": [
        "!wget --no-check-certificate \\\n",
        "  https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip \\\n",
        "  -O /tmp/cats_and_dogs_filtered.zip"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-05-29 10:00:32--  https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip\n",
            "Resolving storage.googleapis.com (storage.googleapis.com)... 108.177.126.128, 2a00:1450:4013:c01::80\n",
            "Connecting to storage.googleapis.com (storage.googleapis.com)|108.177.126.128|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 68606236 (65M) [application/zip]\n",
            "Saving to: ‘/tmp/cats_and_dogs_filtered.zip’\n",
            "\n",
            "\r          /tmp/cats   0%[                    ]       0  --.-KB/s               \r         /tmp/cats_  41%[=======>            ]  27.30M   137MB/s               \r/tmp/cats_and_dogs_ 100%[===================>]  65.43M   226MB/s    in 0.3s    \n",
            "\n",
            "2020-05-29 10:00:32 (226 MB/s) - ‘/tmp/cats_and_dogs_filtered.zip’ saved [68606236/68606236]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7r5jpz4T-wjq",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import os\n",
        "import zipfile"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_4f3_dyJ-_ZK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "local_zip = '/tmp/cats_and_dogs_filtered.zip'\n",
        "zip_ref = zipfile.ZipFile(local_zip, 'r')\n",
        "zip_ref.extractall('/tmp')\n",
        "zip_ref.close()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jwhDSORZ_iza",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\n",
        "train_dir = os.path.join('/tmp/cats_and_dogs_filtered/train')\n",
        "validation_dir = os.path.join('/tmp/cats_and_dogs_filtered/validation')\n",
        "\n",
        "train_cats = os.path.join('/tmp/cats_and_dogs_filtered/train/cats')\n",
        "train_dogs = os.path.join('/tmp/cats_and_dogs_filtered/train/dogs')\n",
        "\n",
        "test_cats = os.path.join('/tmp/cats_and_dogs_filtered/validation/cats')\n",
        "test_dogs = os.path.join('/tmp/cats_and_dogs_filtered/validation/dogs')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qS3qhNF1AFvq",
        "colab_type": "code",
        "outputId": "3d790432-f5f6-4289-e5b3-05dc9b42cb02",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        }
      },
      "source": [
        "\n",
        "print('Total training cats: '+str(len(os.listdir('/tmp/cats_and_dogs_filtered/train/cats'))))\n",
        "print('Total training dogs: '+str(len(os.listdir('/tmp/cats_and_dogs_filtered/train/dogs'))))\n",
        "print('Total validation cats: '+str(len(os.listdir('/tmp/cats_and_dogs_filtered/validation/cats'))))\n",
        "print('Total validation dogs: '+str(len(os.listdir('/tmp/cats_and_dogs_filtered/validation/dogs'))))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Total training cats: 1000\n",
            "Total training dogs: 1000\n",
            "Total validation cats: 500\n",
            "Total validation dogs: 500\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Iu8QjQkOAUO3",
        "colab_type": "code",
        "outputId": "fc19fa1d-3105-44ac-dc86-d199487f6a8a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 459
        }
      },
      "source": [
        "from tensorflow.keras import regularizers\n",
        "\n",
        "model = keras.models.Sequential([\n",
        "                                 keras.layers.Conv2D(16,(3,3),activation='relu',input_shape=(150,150,3)),\n",
        "                                 keras.layers.MaxPooling2D(2,2),\n",
        "                                 keras.layers.Conv2D(32, (3,3), activation='relu'),\n",
        "                                 keras.layers.MaxPooling2D(2,2), \n",
        "                                 keras.layers.Conv2D(64, (3,3), activation='relu'), \n",
        "                                 keras.layers.MaxPooling2D(2,2),\n",
        "                                 # Flatten the results to feed into a DNN\n",
        "                                 keras.layers.Flatten(), \n",
        "                                 # 512 neuron hidden layer\n",
        "                                 keras.layers.Dense(512, activation='relu'), \n",
        "                                 # Only 1 output neuron. It will contain a value from 0-1 where 0 for 1 class ('cats') and 1 for the other ('dogs')\n",
        "                                 keras.layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "model.summary()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_2\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_4 (Conv2D)            (None, 148, 148, 16)      448       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_4 (MaxPooling2 (None, 74, 74, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_5 (Conv2D)            (None, 72, 72, 32)        4640      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_5 (MaxPooling2 (None, 36, 36, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_6 (Conv2D)            (None, 34, 34, 64)        18496     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_6 (MaxPooling2 (None, 17, 17, 64)        0         \n",
            "_________________________________________________________________\n",
            "flatten_2 (Flatten)          (None, 18496)             0         \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 512)               9470464   \n",
            "_________________________________________________________________\n",
            "dense_4 (Dense)              (None, 1)                 513       \n",
            "=================================================================\n",
            "Total params: 9,494,561\n",
            "Trainable params: 9,494,561\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-HYuXDiJC-Pf",
        "colab_type": "code",
        "outputId": "2b6771c3-31a0-4039-8b25-5829e431d7b9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "# All images will be rescaled by 1./255.\n",
        "train_datagen = ImageDataGenerator( rescale = 1.0/255.,\n",
        "                                   rotation_range = 40,\n",
        "                                   width_shift_range = 0.2,\n",
        "                                   height_shift_range = 0.2,\n",
        "                                   shear_range = 0.2,\n",
        "                                   zoom_range = 0.2,\n",
        "                                   horizontal_flip = True,\n",
        "                                   fill_mode = 'nearest')\n",
        "test_datagen  = ImageDataGenerator( rescale = 1.0/255. )\n",
        "\n",
        "# --------------------\n",
        "# Flow training images in batches of 20 using train_datagen generator\n",
        "# --------------------\n",
        "train_generator = train_datagen.flow_from_directory(train_dir,\n",
        "                                                    batch_size=20,\n",
        "                                                    class_mode='binary',\n",
        "                                                    target_size=(150, 150))     \n",
        "# --------------------\n",
        "# Flow validation images in batches of 20 using test_datagen generator\n",
        "# --------------------\n",
        "validation_generator =  test_datagen.flow_from_directory(validation_dir,\n",
        "                                                         batch_size=20,\n",
        "                                                         class_mode  = 'binary',\n",
        "                                                         target_size = (150, 150))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 2000 images belonging to 2 classes.\n",
            "Found 1000 images belonging to 2 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DF97_ROyDG00",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from tensorflow.keras.optimizers import Adam\n",
        "\n",
        "model.compile(loss='binary_crossentropy',\n",
        "              optimizer = 'Adam',\n",
        "              metrics = ['acc']\n",
        "              )"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2a5SvmwmEEjb",
        "colab_type": "code",
        "outputId": "4428e075-47a1-461c-d894-7a9f16313322",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "history = model.fit_generator(\n",
        "    train_generator,\n",
        "    epochs = 10,\n",
        "    validation_data = validation_generator,\n",
        "    verbose = 1\n",
        ")"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "100/100 [==============================] - 21s 209ms/step - loss: 0.5885 - acc: 0.6925 - val_loss: 0.5092 - val_acc: 0.7280\n",
            "Epoch 2/100\n",
            "100/100 [==============================] - 19s 192ms/step - loss: 0.5629 - acc: 0.7160 - val_loss: 0.7875 - val_acc: 0.7280\n",
            "Epoch 3/100\n",
            "100/100 [==============================] - 19s 188ms/step - loss: 0.5536 - acc: 0.7065 - val_loss: 0.7587 - val_acc: 0.7030\n",
            "Epoch 4/100\n",
            "100/100 [==============================] - 19s 190ms/step - loss: 0.5500 - acc: 0.7260 - val_loss: 0.4756 - val_acc: 0.6890\n",
            "Epoch 5/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.5324 - acc: 0.7320 - val_loss: 0.4034 - val_acc: 0.7300\n",
            "Epoch 6/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.5393 - acc: 0.7165 - val_loss: 0.6706 - val_acc: 0.7330\n",
            "Epoch 7/100\n",
            "100/100 [==============================] - 19s 190ms/step - loss: 0.5319 - acc: 0.7300 - val_loss: 0.5264 - val_acc: 0.7380\n",
            "Epoch 8/100\n",
            "100/100 [==============================] - 19s 188ms/step - loss: 0.5220 - acc: 0.7480 - val_loss: 0.2557 - val_acc: 0.7410\n",
            "Epoch 9/100\n",
            "100/100 [==============================] - 19s 188ms/step - loss: 0.5212 - acc: 0.7415 - val_loss: 0.5604 - val_acc: 0.7550\n",
            "Epoch 10/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.5466 - acc: 0.7170 - val_loss: 0.3757 - val_acc: 0.7340\n",
            "Epoch 11/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.5126 - acc: 0.7480 - val_loss: 0.4086 - val_acc: 0.7340\n",
            "Epoch 12/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.5075 - acc: 0.7510 - val_loss: 0.3601 - val_acc: 0.7630\n",
            "Epoch 13/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.5129 - acc: 0.7460 - val_loss: 0.5062 - val_acc: 0.7520\n",
            "Epoch 14/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.5254 - acc: 0.7470 - val_loss: 0.5387 - val_acc: 0.7670\n",
            "Epoch 15/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.4906 - acc: 0.7645 - val_loss: 0.7410 - val_acc: 0.7490\n",
            "Epoch 16/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.5141 - acc: 0.7480 - val_loss: 0.6154 - val_acc: 0.7610\n",
            "Epoch 17/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.5015 - acc: 0.7575 - val_loss: 0.4032 - val_acc: 0.7570\n",
            "Epoch 18/100\n",
            "100/100 [==============================] - 19s 185ms/step - loss: 0.5130 - acc: 0.7455 - val_loss: 0.6569 - val_acc: 0.7750\n",
            "Epoch 19/100\n",
            "100/100 [==============================] - 19s 185ms/step - loss: 0.5021 - acc: 0.7525 - val_loss: 0.6023 - val_acc: 0.7560\n",
            "Epoch 20/100\n",
            "100/100 [==============================] - 19s 185ms/step - loss: 0.4915 - acc: 0.7565 - val_loss: 0.4842 - val_acc: 0.7620\n",
            "Epoch 21/100\n",
            "100/100 [==============================] - 19s 185ms/step - loss: 0.4782 - acc: 0.7810 - val_loss: 0.4346 - val_acc: 0.7630\n",
            "Epoch 22/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.4828 - acc: 0.7645 - val_loss: 0.6277 - val_acc: 0.7030\n",
            "Epoch 23/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.4932 - acc: 0.7625 - val_loss: 0.7574 - val_acc: 0.7410\n",
            "Epoch 24/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.4755 - acc: 0.7720 - val_loss: 0.5331 - val_acc: 0.7520\n",
            "Epoch 25/100\n",
            "100/100 [==============================] - 18s 183ms/step - loss: 0.4517 - acc: 0.7820 - val_loss: 0.5186 - val_acc: 0.7820\n",
            "Epoch 26/100\n",
            "100/100 [==============================] - 18s 184ms/step - loss: 0.4829 - acc: 0.7630 - val_loss: 0.4602 - val_acc: 0.7700\n",
            "Epoch 27/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.4695 - acc: 0.7815 - val_loss: 0.9727 - val_acc: 0.7610\n",
            "Epoch 28/100\n",
            "100/100 [==============================] - 19s 185ms/step - loss: 0.4785 - acc: 0.7725 - val_loss: 0.5611 - val_acc: 0.7430\n",
            "Epoch 29/100\n",
            "100/100 [==============================] - 18s 184ms/step - loss: 0.4591 - acc: 0.7835 - val_loss: 0.5569 - val_acc: 0.7640\n",
            "Epoch 30/100\n",
            "100/100 [==============================] - 18s 184ms/step - loss: 0.4751 - acc: 0.7705 - val_loss: 0.3842 - val_acc: 0.7680\n",
            "Epoch 31/100\n",
            "100/100 [==============================] - 18s 182ms/step - loss: 0.4363 - acc: 0.8025 - val_loss: 0.6922 - val_acc: 0.7510\n",
            "Epoch 32/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.4728 - acc: 0.7710 - val_loss: 0.5352 - val_acc: 0.7620\n",
            "Epoch 33/100\n",
            "100/100 [==============================] - 18s 185ms/step - loss: 0.4589 - acc: 0.7745 - val_loss: 0.3295 - val_acc: 0.7690\n",
            "Epoch 34/100\n",
            "100/100 [==============================] - 18s 184ms/step - loss: 0.4358 - acc: 0.7955 - val_loss: 0.4547 - val_acc: 0.7620\n",
            "Epoch 35/100\n",
            "100/100 [==============================] - 18s 184ms/step - loss: 0.4373 - acc: 0.7925 - val_loss: 0.8501 - val_acc: 0.7770\n",
            "Epoch 36/100\n",
            "100/100 [==============================] - 18s 182ms/step - loss: 0.4499 - acc: 0.7970 - val_loss: 0.5533 - val_acc: 0.7710\n",
            "Epoch 37/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.4389 - acc: 0.7855 - val_loss: 0.7876 - val_acc: 0.7660\n",
            "Epoch 38/100\n",
            "100/100 [==============================] - 18s 183ms/step - loss: 0.4430 - acc: 0.7850 - val_loss: 0.4086 - val_acc: 0.7610\n",
            "Epoch 39/100\n",
            "100/100 [==============================] - 18s 183ms/step - loss: 0.4221 - acc: 0.8095 - val_loss: 0.2458 - val_acc: 0.7440\n",
            "Epoch 40/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.4582 - acc: 0.7895 - val_loss: 0.4611 - val_acc: 0.7560\n",
            "Epoch 41/100\n",
            "100/100 [==============================] - 18s 179ms/step - loss: 0.4444 - acc: 0.7820 - val_loss: 0.5822 - val_acc: 0.7470\n",
            "Epoch 42/100\n",
            "100/100 [==============================] - 18s 181ms/step - loss: 0.4363 - acc: 0.8005 - val_loss: 0.4702 - val_acc: 0.7880\n",
            "Epoch 43/100\n",
            "100/100 [==============================] - 18s 179ms/step - loss: 0.4264 - acc: 0.7965 - val_loss: 0.1837 - val_acc: 0.7630\n",
            "Epoch 44/100\n",
            "100/100 [==============================] - 18s 179ms/step - loss: 0.4357 - acc: 0.8065 - val_loss: 0.3257 - val_acc: 0.7800\n",
            "Epoch 45/100\n",
            "100/100 [==============================] - 18s 179ms/step - loss: 0.4248 - acc: 0.7935 - val_loss: 0.4150 - val_acc: 0.7540\n",
            "Epoch 46/100\n",
            "100/100 [==============================] - 18s 178ms/step - loss: 0.4216 - acc: 0.8070 - val_loss: 0.6053 - val_acc: 0.7780\n",
            "Epoch 47/100\n",
            "100/100 [==============================] - 18s 180ms/step - loss: 0.4021 - acc: 0.8075 - val_loss: 0.7002 - val_acc: 0.7800\n",
            "Epoch 48/100\n",
            "100/100 [==============================] - 18s 178ms/step - loss: 0.3999 - acc: 0.8250 - val_loss: 0.5973 - val_acc: 0.7660\n",
            "Epoch 49/100\n",
            "100/100 [==============================] - 18s 178ms/step - loss: 0.4325 - acc: 0.8005 - val_loss: 0.3712 - val_acc: 0.7860\n",
            "Epoch 50/100\n",
            "100/100 [==============================] - 18s 178ms/step - loss: 0.4039 - acc: 0.8145 - val_loss: 0.4864 - val_acc: 0.7730\n",
            "Epoch 51/100\n",
            "100/100 [==============================] - 18s 177ms/step - loss: 0.3965 - acc: 0.8230 - val_loss: 0.3423 - val_acc: 0.7770\n",
            "Epoch 52/100\n",
            "100/100 [==============================] - 18s 182ms/step - loss: 0.4238 - acc: 0.8020 - val_loss: 0.5090 - val_acc: 0.7730\n",
            "Epoch 53/100\n",
            "100/100 [==============================] - 18s 184ms/step - loss: 0.3996 - acc: 0.8100 - val_loss: 0.4713 - val_acc: 0.7810\n",
            "Epoch 54/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.4044 - acc: 0.8190 - val_loss: 0.2124 - val_acc: 0.7910\n",
            "Epoch 55/100\n",
            "100/100 [==============================] - 19s 188ms/step - loss: 0.4060 - acc: 0.8110 - val_loss: 0.3762 - val_acc: 0.7950\n",
            "Epoch 56/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.4055 - acc: 0.8185 - val_loss: 0.4514 - val_acc: 0.7840\n",
            "Epoch 57/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.4011 - acc: 0.8105 - val_loss: 0.3865 - val_acc: 0.7820\n",
            "Epoch 58/100\n",
            "100/100 [==============================] - 18s 181ms/step - loss: 0.3916 - acc: 0.8215 - val_loss: 0.4547 - val_acc: 0.7820\n",
            "Epoch 59/100\n",
            "100/100 [==============================] - 18s 182ms/step - loss: 0.4023 - acc: 0.8150 - val_loss: 0.9321 - val_acc: 0.7830\n",
            "Epoch 60/100\n",
            "100/100 [==============================] - 18s 183ms/step - loss: 0.3798 - acc: 0.8320 - val_loss: 0.5169 - val_acc: 0.7800\n",
            "Epoch 61/100\n",
            "100/100 [==============================] - 18s 184ms/step - loss: 0.3722 - acc: 0.8330 - val_loss: 0.8443 - val_acc: 0.7800\n",
            "Epoch 62/100\n",
            "100/100 [==============================] - 19s 185ms/step - loss: 0.3884 - acc: 0.8240 - val_loss: 0.6126 - val_acc: 0.7930\n",
            "Epoch 63/100\n",
            "100/100 [==============================] - 18s 183ms/step - loss: 0.3797 - acc: 0.8230 - val_loss: 0.5496 - val_acc: 0.8010\n",
            "Epoch 64/100\n",
            "100/100 [==============================] - 18s 184ms/step - loss: 0.3651 - acc: 0.8440 - val_loss: 0.3519 - val_acc: 0.7910\n",
            "Epoch 65/100\n",
            "100/100 [==============================] - 18s 184ms/step - loss: 0.3670 - acc: 0.8350 - val_loss: 0.3024 - val_acc: 0.7900\n",
            "Epoch 66/100\n",
            "100/100 [==============================] - 18s 184ms/step - loss: 0.3790 - acc: 0.8355 - val_loss: 0.4731 - val_acc: 0.8020\n",
            "Epoch 67/100\n",
            "100/100 [==============================] - 19s 188ms/step - loss: 0.3631 - acc: 0.8325 - val_loss: 0.5437 - val_acc: 0.7740\n",
            "Epoch 68/100\n",
            "100/100 [==============================] - 18s 185ms/step - loss: 0.3547 - acc: 0.8405 - val_loss: 0.4093 - val_acc: 0.7910\n",
            "Epoch 69/100\n",
            "100/100 [==============================] - 19s 185ms/step - loss: 0.3448 - acc: 0.8460 - val_loss: 0.4637 - val_acc: 0.7890\n",
            "Epoch 70/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3534 - acc: 0.8450 - val_loss: 0.1025 - val_acc: 0.7980\n",
            "Epoch 71/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3613 - acc: 0.8330 - val_loss: 0.5481 - val_acc: 0.7780\n",
            "Epoch 72/100\n",
            "100/100 [==============================] - 19s 188ms/step - loss: 0.3798 - acc: 0.8255 - val_loss: 0.7129 - val_acc: 0.7880\n",
            "Epoch 73/100\n",
            "100/100 [==============================] - 19s 188ms/step - loss: 0.3511 - acc: 0.8400 - val_loss: 0.4183 - val_acc: 0.8020\n",
            "Epoch 74/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3645 - acc: 0.8490 - val_loss: 0.3877 - val_acc: 0.8010\n",
            "Epoch 75/100\n",
            "100/100 [==============================] - 18s 185ms/step - loss: 0.3414 - acc: 0.8415 - val_loss: 0.6658 - val_acc: 0.7970\n",
            "Epoch 76/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3336 - acc: 0.8510 - val_loss: 0.5876 - val_acc: 0.7980\n",
            "Epoch 77/100\n",
            "100/100 [==============================] - 19s 189ms/step - loss: 0.3681 - acc: 0.8360 - val_loss: 0.5010 - val_acc: 0.7870\n",
            "Epoch 78/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3343 - acc: 0.8495 - val_loss: 0.6631 - val_acc: 0.7960\n",
            "Epoch 79/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.3499 - acc: 0.8455 - val_loss: 0.3098 - val_acc: 0.7840\n",
            "Epoch 80/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.3460 - acc: 0.8560 - val_loss: 0.6685 - val_acc: 0.7910\n",
            "Epoch 81/100\n",
            "100/100 [==============================] - 19s 185ms/step - loss: 0.3163 - acc: 0.8660 - val_loss: 0.3634 - val_acc: 0.7970\n",
            "Epoch 82/100\n",
            "100/100 [==============================] - 19s 188ms/step - loss: 0.3498 - acc: 0.8435 - val_loss: 0.2552 - val_acc: 0.8030\n",
            "Epoch 83/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3667 - acc: 0.8425 - val_loss: 0.6501 - val_acc: 0.7700\n",
            "Epoch 84/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3290 - acc: 0.8495 - val_loss: 0.3347 - val_acc: 0.7980\n",
            "Epoch 85/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3262 - acc: 0.8630 - val_loss: 0.5421 - val_acc: 0.7930\n",
            "Epoch 86/100\n",
            "100/100 [==============================] - 19s 187ms/step - loss: 0.3350 - acc: 0.8465 - val_loss: 0.2222 - val_acc: 0.7920\n",
            "Epoch 87/100\n",
            "100/100 [==============================] - 19s 189ms/step - loss: 0.3207 - acc: 0.8630 - val_loss: 0.3365 - val_acc: 0.7990\n",
            "Epoch 88/100\n",
            "100/100 [==============================] - 19s 190ms/step - loss: 0.3368 - acc: 0.8550 - val_loss: 0.4335 - val_acc: 0.7850\n",
            "Epoch 89/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3063 - acc: 0.8610 - val_loss: 0.3036 - val_acc: 0.7910\n",
            "Epoch 90/100\n",
            "100/100 [==============================] - 18s 185ms/step - loss: 0.3071 - acc: 0.8700 - val_loss: 0.1812 - val_acc: 0.8080\n",
            "Epoch 91/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3128 - acc: 0.8595 - val_loss: 0.3465 - val_acc: 0.8070\n",
            "Epoch 92/100\n",
            "100/100 [==============================] - 19s 193ms/step - loss: 0.3301 - acc: 0.8555 - val_loss: 0.3703 - val_acc: 0.8040\n",
            "Epoch 93/100\n",
            "100/100 [==============================] - 19s 191ms/step - loss: 0.3081 - acc: 0.8700 - val_loss: 0.3984 - val_acc: 0.8080\n",
            "Epoch 94/100\n",
            "100/100 [==============================] - 19s 191ms/step - loss: 0.3067 - acc: 0.8750 - val_loss: 0.9350 - val_acc: 0.8060\n",
            "Epoch 95/100\n",
            "100/100 [==============================] - 19s 190ms/step - loss: 0.2944 - acc: 0.8765 - val_loss: 0.2333 - val_acc: 0.8010\n",
            "Epoch 96/100\n",
            "100/100 [==============================] - 19s 185ms/step - loss: 0.2884 - acc: 0.8795 - val_loss: 0.2109 - val_acc: 0.7930\n",
            "Epoch 97/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.2927 - acc: 0.8760 - val_loss: 0.3762 - val_acc: 0.7810\n",
            "Epoch 98/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.2907 - acc: 0.8650 - val_loss: 0.2782 - val_acc: 0.8040\n",
            "Epoch 99/100\n",
            "100/100 [==============================] - 19s 186ms/step - loss: 0.3013 - acc: 0.8655 - val_loss: 0.5497 - val_acc: 0.7900\n",
            "Epoch 100/100\n",
            "100/100 [==============================] - 19s 185ms/step - loss: 0.3068 - acc: 0.8720 - val_loss: 1.5111 - val_acc: 0.7670\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FAADtzVHOpj2",
        "colab_type": "code",
        "outputId": "0241ba26-7d36-4a6c-e817-04e8e87506d4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        }
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "acc = history.history['acc']\n",
        "val_acc = history.history['val_acc']\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs = range(len(acc))\n",
        "\n",
        "plt.plot(epochs, acc, 'r', label='Training accuracy')\n",
        "plt.plot(epochs, val_acc, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\n",
        "\n",
        "plt.figure()\n",
        "\n",
        "plt.plot(epochs, loss, 'r', label='Training Loss')\n",
        "plt.plot(epochs, val_loss, 'b', label='Validation Loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ACpWrKWHXAQj",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}