VGG19.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.6.10"
    },
    "colab": {
      "name": "cnn.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "YGUitmvt8LBb",
        "colab_type": "code",
        "outputId": "ff8f5555-d72c-420c-92f0-5b7060eec35c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 55
        }
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-nL2qOf96Uwu",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#\n",
        "# **************************************************************************************************************************************************\n",
        "#\tLibraries\n",
        "# **************************************************************************************************************************************************\n",
        "#\n",
        "\n",
        "from keras import optimizers\n",
        "from keras import applications\n",
        "from keras.models import Model\n",
        "from keras.layers import Flatten, Dense\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "from keras.utils.np_utils import to_categorical\n",
        "from keras.callbacks import EarlyStopping, ModelCheckpoint\n",
        "from math import ceil\n",
        "\n",
        "import datetime\n",
        "import json\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Bnbx-fSDJ5-E",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#\n",
        "# **************************************************************************************************************************************************\n",
        "#\tGlobal Variables\n",
        "# **************************************************************************************************************************************************\n",
        "#\n",
        "\n",
        "# saved filename of model\n",
        "model_filename = '/content/drive/My Drive/FYP/Project/model.h5'\n",
        "\n",
        "# saved filename of model weights\n",
        "weights_filename = '/content/drive/My Drive/FYP/Project/model_weights.h5'\n",
        "modelcheckpoint_filename = '/content/drive/My Drive/FYP/Project/model_checkpoint.h5'\n",
        "\n",
        "# saved filename of result csv\n",
        "results_filename = '/content/drive/My Drive/FYP/Project/results.csv'\n",
        "\n",
        "# directories of dataset\n",
        "train_data_dir = '/content/drive/My Drive/FYP/Project/data/train'\n",
        "# validation_data_dir = '/content/drive/My Drive/FYP/Project/data/validation'\n",
        "test_data_dir = '/content/drive/My Drive/FYP/Project/data/test'\n",
        "output_dir = '/content/drive/My Drive/FYP/Project/output'\n",
        "\n",
        "# dimensions of the architecture\n",
        "img_width, img_height = 224, 224\n",
        "\n",
        "# number of channels for the architecture\n",
        "channels = 3\n",
        "\n",
        "# number of epochs\n",
        "epochs = 34\n",
        "\n",
        "# batch sizes\n",
        "training_batch_size = 2\n",
        "validation_batch_size = 1\n",
        "\n",
        "# learning rate\n",
        "learning_rate = 0.0001\n",
        "\n",
        "# class mode\n",
        "class_mode = 'categorical'\n",
        "\n",
        "# total number of classes\n",
        "total_classes = 2"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9RmONmQlqQJ0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#\n",
        "# **************************************************************************************************************************************************\n",
        "#\tFunctions\n",
        "# **************************************************************************************************************************************************\n",
        "#\n",
        "\n",
        "def summary():\n",
        "\t# model summary\n",
        "\tvgg19.summary()\n",
        "\n",
        "\t# graphing our training and validation\n",
        "\tacc = history.history['acc']\n",
        "\tval_acc = history.history['val_acc']\n",
        "\tloss = history.history['loss']\n",
        "\tval_loss = history.history['val_loss']\n",
        "\tepochs = range(len(acc))\n",
        "\tplt.plot(epochs, acc, 'r', label = 'Training Acc')\n",
        "\tplt.plot(epochs, val_acc, 'b', label = 'Validation Acc')\n",
        "\tplt.title('Training and Validation Accuracy')\n",
        "\tplt.ylabel('Accuracy')\n",
        "\tplt.xlabel('Epoch')\n",
        "\tplt.legend()\n",
        "\tplt.figure()\n",
        "\tplt.plot(epochs, loss, 'r', label = 'Training Loss')\n",
        "\tplt.plot(epochs, val_loss, 'b', label = 'Validation Loss')\n",
        "\tplt.title('Training and Validation Loss')\n",
        "\tplt.ylabel('Loss')\n",
        "\tplt.xlabel('Epoch')\n",
        "\tplt.legend()\n",
        "\tplt.show()\n",
        "\n",
        "def saveModel():\n",
        "\tmodel.save(model_filename)\n",
        "\tmodel.save_weights(weights_filename)\n",
        "\n",
        "\tmodel_json = model.to_json()\n",
        "\twith open(\"model_in_json.json\", \"w\") as json_file:\n",
        "\t\tjson.dump(model_json, json_file)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V8elbJX-6Uw5",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#\n",
        "# **************************************************************************************************************************************************\n",
        "#\tMain Code\n",
        "# **************************************************************************************************************************************************\n",
        "#\n",
        "\n",
        "# load model\n",
        "model = applications.VGG19(include_top = False, weights = 'imagenet', classes = total_classes, input_shape = (img_width, img_height, channels))\n",
        "flat = Flatten()(model.layers[-1].output)\n",
        "output = Dense(total_classes, activation = \"softmax\")(flat)\n",
        "vgg19 = Model(inputs = model.inputs, outputs = output)\n",
        "\n",
        "train_datagen = ImageDataGenerator(rescale = 1. / 255,\n",
        "                                    rotation_range = 90,\n",
        "                                    brightness_range = [0.2, 1.0],\n",
        "                                    shear_range = 0.2,\n",
        "                                    zoom_range = 0.3,\n",
        "                                    width_shift_range = 1.0,\n",
        "                                    height_shift_range = 1.0,\n",
        "                                    horizontal_flip = True,\n",
        "                                    vertical_flip = True,\n",
        "                                    validation_split = 0.4)\n",
        "\n",
        "# validation_datagen = ImageDataGenerator(rescale = 1. / 255)\n",
        "\n",
        "# current time\n",
        "start = datetime.datetime.now()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aRufvoBe6Uw8",
        "colab_type": "code",
        "outputId": "662c9e38-db3e-4082-dade-292620cfd990",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "# training data\n",
        "training_set = train_datagen.flow_from_directory(\n",
        "                                \t\ttrain_data_dir,\n",
        "                                \t\ttarget_size = (img_width, img_height),\n",
        "                                \t\tbatch_size = training_batch_size,\n",
        "                                \t\tclass_mode = class_mode,\n",
        "\t\t\t\t\t\t\t\t\t\t#save_to_dir = output_dir,\n",
        "\t\t\t\t\t\t\t\t\t\t#save_prefix = 'Aug',\n",
        "\t\t\t\t\t\t\t\t\t\tsubset = \"training\",\n",
        "\t\t\t\t\t\t\t\t\t\tseed = 42,\n",
        "                                \t\tshuffle = True)\n",
        "\n",
        "nb_train_samples = len(training_set.filenames)"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 352 images belonging to 2 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rYqz-nQ36UxB",
        "colab_type": "code",
        "outputId": "25b271e5-445a-4387-cf82-caa5dba85753",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "# validation data\n",
        "validation_set = train_datagen.flow_from_directory(\n",
        "                                  \t\ttrain_data_dir,\n",
        "                                  \t\ttarget_size = (img_width, img_height),\n",
        "                                  \t\tbatch_size = validation_batch_size,\n",
        "                                  \t\tclass_mode = class_mode,\n",
        "\t\t\t\t\t\t\t\t\t\tsubset = \"validation\",\n",
        "\t\t\t\t\t\t\t\t\t\tseed = 42,\n",
        "                                  \t\tshuffle = True)\n",
        "\n",
        "nb_validation_samples = len(validation_set.filenames)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 232 images belonging to 2 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": false,
        "id": "XfOVOtLj6UxN",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\t\t\t\tvgg19.compile(loss = 'categorical_crossentropy',\n",
        "\t\t\t\toptimizer = optimizers.SGD(lr = learning_rate),\n",
        "\t\t\t\tmetrics = ['acc'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PKmIEdJl6UxS",
        "colab_type": "code",
        "outputId": "07dbf5f5-a3e2-41d8-8876-ef92b65b99e4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "# es = EarlyStopping(monitor = 'val_loss',\n",
        "#                     min_delta = 1e-5,\n",
        "#                     patience = 5,\n",
        "#                     verbose = 1,\n",
        "#                     mode = 'min',\n",
        "#                     restore_best_weights = True)\n",
        "\n",
        "mc = ModelCheckpoint(modelcheckpoint_filename,\n",
        "                        monitor = 'val_loss',\n",
        "                        mode = 'min',\n",
        "                        save_weights_only = True,\n",
        "                        save_best_only = True)\n",
        "\n",
        "history = vgg19.fit_generator(training_set,\n",
        "                                steps_per_epoch = nb_train_samples / training_batch_size,\n",
        "                                epochs = epochs,\n",
        "                              #remember to put back afteer confiq 2 callbacks = [es,mc],\n",
        "                                callbacks = [mc],\n",
        "                                validation_data = validation_set,\n",
        "                                validation_steps = nb_validation_samples / validation_batch_size)"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/34\n",
            "176/176 [==============================] - 19s 106ms/step - loss: 0.7689 - acc: 0.4631 - val_loss: 0.7342 - val_acc: 0.4871\n",
            "Epoch 2/34\n",
            "176/176 [==============================] - 17s 98ms/step - loss: 0.7545 - acc: 0.5341 - val_loss: 0.7321 - val_acc: 0.4957\n",
            "Epoch 3/34\n",
            "176/176 [==============================] - 17s 97ms/step - loss: 0.7418 - acc: 0.5028 - val_loss: 0.6887 - val_acc: 0.5216\n",
            "Epoch 4/34\n",
            "176/176 [==============================] - 17s 95ms/step - loss: 0.7395 - acc: 0.5284 - val_loss: 0.6956 - val_acc: 0.5216\n",
            "Epoch 5/34\n",
            "176/176 [==============================] - 17s 97ms/step - loss: 0.7291 - acc: 0.5597 - val_loss: 0.7705 - val_acc: 0.5086\n",
            "Epoch 6/34\n",
            "176/176 [==============================] - 17s 97ms/step - loss: 0.7583 - acc: 0.4915 - val_loss: 0.6876 - val_acc: 0.5647\n",
            "Epoch 7/34\n",
            "176/176 [==============================] - 17s 97ms/step - loss: 0.7102 - acc: 0.5625 - val_loss: 0.6887 - val_acc: 0.5259\n",
            "Epoch 8/34\n",
            "176/176 [==============================] - 17s 95ms/step - loss: 0.7171 - acc: 0.5483 - val_loss: 0.6892 - val_acc: 0.5259\n",
            "Epoch 9/34\n",
            "176/176 [==============================] - 16s 93ms/step - loss: 0.7070 - acc: 0.5682 - val_loss: 0.6777 - val_acc: 0.5948\n",
            "Epoch 10/34\n",
            "176/176 [==============================] - 16s 93ms/step - loss: 0.6916 - acc: 0.5938 - val_loss: 0.6959 - val_acc: 0.5259\n",
            "Epoch 11/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.6798 - acc: 0.5909 - val_loss: 0.7158 - val_acc: 0.5302\n",
            "Epoch 12/34\n",
            "176/176 [==============================] - 16s 90ms/step - loss: 0.6997 - acc: 0.5852 - val_loss: 0.6815 - val_acc: 0.5345\n",
            "Epoch 13/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.6779 - acc: 0.5966 - val_loss: 0.6776 - val_acc: 0.5819\n",
            "Epoch 14/34\n",
            "176/176 [==============================] - 16s 93ms/step - loss: 0.6748 - acc: 0.5938 - val_loss: 0.7927 - val_acc: 0.5086\n",
            "Epoch 15/34\n",
            "176/176 [==============================] - 17s 94ms/step - loss: 0.6725 - acc: 0.6080 - val_loss: 0.6962 - val_acc: 0.5647\n",
            "Epoch 16/34\n",
            "176/176 [==============================] - 16s 93ms/step - loss: 0.6572 - acc: 0.6278 - val_loss: 0.6787 - val_acc: 0.5603\n",
            "Epoch 17/34\n",
            "176/176 [==============================] - 16s 93ms/step - loss: 0.6819 - acc: 0.5966 - val_loss: 0.6798 - val_acc: 0.5388\n",
            "Epoch 18/34\n",
            "176/176 [==============================] - 16s 93ms/step - loss: 0.6385 - acc: 0.6534 - val_loss: 0.6799 - val_acc: 0.5948\n",
            "Epoch 19/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.6477 - acc: 0.6392 - val_loss: 0.6826 - val_acc: 0.5517\n",
            "Epoch 20/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.6184 - acc: 0.6705 - val_loss: 0.6719 - val_acc: 0.6121\n",
            "Epoch 21/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.6386 - acc: 0.6364 - val_loss: 0.6413 - val_acc: 0.6164\n",
            "Epoch 22/34\n",
            "176/176 [==============================] - 16s 91ms/step - loss: 0.5990 - acc: 0.6676 - val_loss: 0.6920 - val_acc: 0.5603\n",
            "Epoch 23/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.6358 - acc: 0.6506 - val_loss: 0.7119 - val_acc: 0.5129\n",
            "Epoch 24/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.6096 - acc: 0.7045 - val_loss: 0.6661 - val_acc: 0.5991\n",
            "Epoch 25/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.5699 - acc: 0.7131 - val_loss: 0.6801 - val_acc: 0.6121\n",
            "Epoch 26/34\n",
            "176/176 [==============================] - 16s 91ms/step - loss: 0.5728 - acc: 0.7017 - val_loss: 0.6729 - val_acc: 0.5905\n",
            "Epoch 27/34\n",
            "176/176 [==============================] - 16s 91ms/step - loss: 0.5872 - acc: 0.6989 - val_loss: 0.7750 - val_acc: 0.5345\n",
            "Epoch 28/34\n",
            "176/176 [==============================] - 16s 94ms/step - loss: 0.5832 - acc: 0.7074 - val_loss: 0.7333 - val_acc: 0.5474\n",
            "Epoch 29/34\n",
            "176/176 [==============================] - 16s 91ms/step - loss: 0.6091 - acc: 0.6847 - val_loss: 0.6432 - val_acc: 0.6336\n",
            "Epoch 30/34\n",
            "176/176 [==============================] - 16s 91ms/step - loss: 0.5782 - acc: 0.6705 - val_loss: 0.7021 - val_acc: 0.5948\n",
            "Epoch 31/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.5752 - acc: 0.7244 - val_loss: 0.6825 - val_acc: 0.5819\n",
            "Epoch 32/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.5600 - acc: 0.7131 - val_loss: 0.6698 - val_acc: 0.6034\n",
            "Epoch 33/34\n",
            "176/176 [==============================] - 16s 92ms/step - loss: 0.5468 - acc: 0.7131 - val_loss: 0.7640 - val_acc: 0.5776\n",
            "Epoch 34/34\n",
            "176/176 [==============================] - 16s 91ms/step - loss: 0.5417 - acc: 0.7188 - val_loss: 0.6902 - val_acc: 0.6034\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2rxkOJ1uKh6x",
        "colab_type": "code",
        "outputId": "9503e19e-b5fa-4b79-f864-644360104e6f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "summary()\n",
        "saveModel()"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"model_3\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "input_3 (InputLayer)         (None, 224, 224, 3)       0         \n",
            "_________________________________________________________________\n",
            "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
            "_________________________________________________________________\n",
            "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
            "_________________________________________________________________\n",
            "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
            "_________________________________________________________________\n",
            "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
            "_________________________________________________________________\n",
            "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
            "_________________________________________________________________\n",
            "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
            "_________________________________________________________________\n",
            "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
            "_________________________________________________________________\n",
            "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_conv4 (Conv2D)        (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
            "_________________________________________________________________\n",
            "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
            "_________________________________________________________________\n",
            "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_conv4 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_conv4 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
            "_________________________________________________________________\n",
            "flatten_3 (Flatten)          (None, 25088)             0         \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 2)                 50178     \n",
            "=================================================================\n",
            "Total params: 20,074,562\n",
            "Trainable params: 20,074,562\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKRAREVHrf0Ax6eJAgFI3AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VRCgdcuEp7cg",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-iiItUupKjD2",
        "colab_type": "code",
        "outputId": "ffc16ca8-eaf1-4c8d-b71c-a714765cb7cd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "  (eval_loss, eval_accuracy) = vgg19.evaluate_generator(\n",
        " validation_set, steps = nb_validation_samples, verbose = 1)"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "232/232 [==============================] - 5s 21ms/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uWCzxZFPKj9s",
        "colab_type": "code",
        "outputId": "cd8348f7-9b83-4247-b0bf-4ac9ea800ee9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 70
        }
      },
      "source": [
        "print(\"[INFO] Accuracy: {:.2f}%\".format(eval_accuracy * 100))\n",
        "print(\"[INFO] Loss: {}\".format(eval_loss))\n",
        "end = datetime.datetime.now()\n",
        "elapsed = end - start\n",
        "print('[INFO] Time Elapsed: ', elapsed)"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[INFO] Accuracy: 62.50%\n",
            "[INFO] Loss: 0.670623984444758\n",
            "[INFO] Time Elapsed:  0:09:34.726727\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZrjRKItL8kc2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NvQg4XHJT7gf",
        "colab_type": "text"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8XiSCTTq5jbJ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "2ab7d628-436e-4d45-eb1e-2bde6f6cddd6"
      },
      "source": [
        "# test data\n",
        "test_datagen = ImageDataGenerator()\n",
        "test_set = test_datagen.flow_from_directory(\n",
        "                                  \t\ttest_data_dir,\n",
        "                                  \t\ttarget_size = (img_width, img_height),\n",
        "                                  \t\tbatch_size = 1,\n",
        "                                  \t\tclass_mode = class_mode,\n",
        "\t\t\t\t\t\t\t\t\t\tseed = 42,\n",
        "                                  \t\tshuffle = True)\n",
        "\n",
        "step_size_test = test_set.n // test_set.batch_size\n",
        "test_set.reset()\n",
        "\n",
        "pred = vgg19.predict_generator(test_set,\n",
        "                                steps = step_size_test,\n",
        "                                verbose = 1)\n",
        "\n",
        "predicted_class_indices = np.argmax(pred, axis = 1)\n",
        "labels = (test_set.class_indices)\n",
        "\n",
        "labels = dict((v, k) for k, v in labels.items())\n",
        "predictions = [labels[k] for k in predicted_class_indices]\n",
        "\n",
        "filenames = test_set.filenames\n",
        "results = pd.DataFrame({\"Filename\": filenames,\n",
        "                      \"Prediction\": predictions})\n",
        "results.to_csv(results_filename, index = False)"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 352 images belonging to 2 classes.\n",
            "352/352 [==============================] - 4s 13ms/step\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}