Deep Learning CNN Santa Datasets.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Bonus_Assignment15_RickyGunawan.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "bpQAQHwTWXiZ"
      },
      "source": [
        "zip_path = '/content/drive/My\\ Drive/BCML/datasets/DeepLearning/santa-dataset.zip'\n",
        "\n",
        "!cp {zip_path} /content/\n",
        "\n",
        "!cd /content/\n",
        "\n",
        "!unzip -q /content/santa-dataset.zip -d /content\n",
        "\n",
        "!rm /content/santa-dataset.zip"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JgSpU5ZJXBme"
      },
      "source": [
        "def plot_accuracy(history):\n",
        "    plt.plot(history.history['accuracy'], label='accuracy')\n",
        "    plt.plot(history.history['val_accuracy'], label='val_accuracy')\n",
        "    plt.xlabel('Epoch')\n",
        "    plt.ylabel('accuracy')\n",
        "    plt.legend()\n",
        "    plt.grid(True)\n",
        "\n",
        "def plot_loss(history):\n",
        "    plt.plot(history.history['loss'], label='loss')\n",
        "    plt.plot(history.history['val_loss'], label='val_loss')\n",
        "    plt.xlabel('Epoch')\n",
        "    plt.ylabel('Loss')\n",
        "    plt.legend()\n",
        "    plt.grid(True)"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0fDjjdhIXG3E"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import os\n",
        "\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras.utils import to_categorical\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Conv2D\n",
        "from tensorflow.keras.layers import MaxPooling2D\n",
        "from tensorflow.keras.layers import Dense\n",
        "from tensorflow.keras.layers import Flatten\n",
        "from tensorflow.keras.optimizers import SGD , RMSprop\n",
        "from tensorflow.keras.layers import BatchNormalization\n",
        "from tensorflow.keras.layers import Dropout"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ObPuYkN8ql-O"
      },
      "source": [
        "# **Base Model**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SlSM8yCJwSXQ",
        "outputId": "aa79443b-2a96-45a4-966f-16143942a5db",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 323
        }
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Conv2D(32, (3, 3), activation='relu', kernel_initializer='he_uniform', padding='same', input_shape=(200, 200, 3)))\n",
        "model.add(MaxPooling2D((2, 2)))\n",
        "\n",
        "model.add(Flatten())\n",
        "\n",
        "model.add(Dense(128, activation='relu', kernel_initializer='he_uniform'))\n",
        "model.add(Dense(1, activation='sigmoid'))\n",
        "\n",
        "opt = SGD(lr=0.001, momentum=0.9)\n",
        "\n",
        "model.compile(optimizer=opt, loss='binary_crossentropy', metrics=['accuracy'])\n",
        "\n",
        "model.summary()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 200, 200, 32)      896       \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 100, 100, 32)      0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 320000)            0         \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 128)               40960128  \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 1)                 129       \n",
            "=================================================================\n",
            "Total params: 40,961,153\n",
            "Trainable params: 40,961,153\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3OUmunEpzExb",
        "outputId": "df41e8a2-4edf-4e8b-aaa8-7b8fafd0dbac",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "dataset_dir = '/content/santa-dataset'\n",
        "\n",
        "datagen = ImageDataGenerator(rescale=1.0/255.0)\n",
        "\n",
        "train_iterator = datagen.flow_from_directory(os.path.join(dataset_dir, 'train'), class_mode='binary', batch_size=128, target_size=(200, 200))\n",
        "test_iterator = datagen.flow_from_directory(os.path.join(dataset_dir, 'test'), class_mode='binary', batch_size=128, target_size=(200, 200))"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 662 images belonging to 2 classes.\n",
            "Found 260 images belonging to 2 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_o0w2sa3Ximv",
        "outputId": "09db1777-ee3c-4bc0-fba8-60536dd30001",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "history=model.fit(train_iterator, steps_per_epoch=len(train_iterator), validation_data=test_iterator, validation_steps=len(test_iterator), epochs=75)"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 1.0128 - accuracy: 0.5227 - val_loss: 0.6827 - val_accuracy: 0.5269\n",
            "Epoch 2/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.6627 - accuracy: 0.5423 - val_loss: 0.6899 - val_accuracy: 0.5923\n",
            "Epoch 3/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.6467 - accuracy: 0.6420 - val_loss: 0.8098 - val_accuracy: 0.5192\n",
            "Epoch 4/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.6093 - accuracy: 0.6752 - val_loss: 0.5766 - val_accuracy: 0.7115\n",
            "Epoch 5/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.4890 - accuracy: 0.7900 - val_loss: 0.5719 - val_accuracy: 0.7231\n",
            "Epoch 6/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.5377 - accuracy: 0.7326 - val_loss: 0.5435 - val_accuracy: 0.7231\n",
            "Epoch 7/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.4464 - accuracy: 0.7749 - val_loss: 0.6661 - val_accuracy: 0.6308\n",
            "Epoch 8/75\n",
            "6/6 [==============================] - 14s 2s/step - loss: 0.4410 - accuracy: 0.8036 - val_loss: 0.6325 - val_accuracy: 0.6654\n",
            "Epoch 9/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.4053 - accuracy: 0.8127 - val_loss: 0.4652 - val_accuracy: 0.8115\n",
            "Epoch 10/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.3313 - accuracy: 0.8852 - val_loss: 0.4934 - val_accuracy: 0.7654\n",
            "Epoch 11/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.3398 - accuracy: 0.8640 - val_loss: 0.4604 - val_accuracy: 0.8192\n",
            "Epoch 12/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.2916 - accuracy: 0.8973 - val_loss: 0.3953 - val_accuracy: 0.8538\n",
            "Epoch 13/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.2619 - accuracy: 0.9184 - val_loss: 0.4448 - val_accuracy: 0.8231\n",
            "Epoch 14/75\n",
            "6/6 [==============================] - 14s 2s/step - loss: 0.2601 - accuracy: 0.9124 - val_loss: 0.4438 - val_accuracy: 0.8192\n",
            "Epoch 15/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.2384 - accuracy: 0.9230 - val_loss: 0.3651 - val_accuracy: 0.8577\n",
            "Epoch 16/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.1966 - accuracy: 0.9411 - val_loss: 0.3583 - val_accuracy: 0.8731\n",
            "Epoch 17/75\n",
            "6/6 [==============================] - 11s 2s/step - loss: 0.1998 - accuracy: 0.9350 - val_loss: 0.3590 - val_accuracy: 0.8769\n",
            "Epoch 18/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.1760 - accuracy: 0.9471 - val_loss: 0.3720 - val_accuracy: 0.8731\n",
            "Epoch 19/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.1947 - accuracy: 0.9335 - val_loss: 0.3682 - val_accuracy: 0.8808\n",
            "Epoch 20/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.1969 - accuracy: 0.9230 - val_loss: 0.3869 - val_accuracy: 0.8692\n",
            "Epoch 21/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.1674 - accuracy: 0.9486 - val_loss: 0.5517 - val_accuracy: 0.7577\n",
            "Epoch 22/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.2470 - accuracy: 0.8897 - val_loss: 0.3616 - val_accuracy: 0.8808\n",
            "Epoch 23/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.1826 - accuracy: 0.9215 - val_loss: 0.5486 - val_accuracy: 0.7692\n",
            "Epoch 24/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.1803 - accuracy: 0.9335 - val_loss: 0.4181 - val_accuracy: 0.8577\n",
            "Epoch 25/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.1335 - accuracy: 0.9577 - val_loss: 0.3509 - val_accuracy: 0.8923\n",
            "Epoch 26/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.1028 - accuracy: 0.9668 - val_loss: 0.3476 - val_accuracy: 0.8962\n",
            "Epoch 27/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0965 - accuracy: 0.9683 - val_loss: 0.3662 - val_accuracy: 0.8808\n",
            "Epoch 28/75\n",
            "6/6 [==============================] - 14s 2s/step - loss: 0.1101 - accuracy: 0.9653 - val_loss: 0.3380 - val_accuracy: 0.8923\n",
            "Epoch 29/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0896 - accuracy: 0.9758 - val_loss: 0.3494 - val_accuracy: 0.8962\n",
            "Epoch 30/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0799 - accuracy: 0.9819 - val_loss: 0.3406 - val_accuracy: 0.8923\n",
            "Epoch 31/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0792 - accuracy: 0.9743 - val_loss: 0.3646 - val_accuracy: 0.8769\n",
            "Epoch 32/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0732 - accuracy: 0.9849 - val_loss: 0.3455 - val_accuracy: 0.8846\n",
            "Epoch 33/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0629 - accuracy: 0.9894 - val_loss: 0.3416 - val_accuracy: 0.8962\n",
            "Epoch 34/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0606 - accuracy: 0.9909 - val_loss: 0.3450 - val_accuracy: 0.8962\n",
            "Epoch 35/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0570 - accuracy: 0.9940 - val_loss: 0.3461 - val_accuracy: 0.8923\n",
            "Epoch 36/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0548 - accuracy: 0.9924 - val_loss: 0.3496 - val_accuracy: 0.8962\n",
            "Epoch 37/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0505 - accuracy: 0.9924 - val_loss: 0.3545 - val_accuracy: 0.8885\n",
            "Epoch 38/75\n",
            "6/6 [==============================] - 14s 2s/step - loss: 0.0479 - accuracy: 0.9955 - val_loss: 0.3495 - val_accuracy: 0.8923\n",
            "Epoch 39/75\n",
            "6/6 [==============================] - 11s 2s/step - loss: 0.0461 - accuracy: 0.9970 - val_loss: 0.3508 - val_accuracy: 0.9000\n",
            "Epoch 40/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0507 - accuracy: 0.9924 - val_loss: 0.3500 - val_accuracy: 0.8923\n",
            "Epoch 41/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0552 - accuracy: 0.9909 - val_loss: 0.3590 - val_accuracy: 0.8885\n",
            "Epoch 42/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0456 - accuracy: 0.9924 - val_loss: 0.3546 - val_accuracy: 0.8885\n",
            "Epoch 43/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0372 - accuracy: 0.9985 - val_loss: 0.3873 - val_accuracy: 0.8731\n",
            "Epoch 44/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0376 - accuracy: 0.9955 - val_loss: 0.3638 - val_accuracy: 0.8885\n",
            "Epoch 45/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0337 - accuracy: 0.9970 - val_loss: 0.3610 - val_accuracy: 0.8962\n",
            "Epoch 46/75\n",
            "6/6 [==============================] - 14s 2s/step - loss: 0.0353 - accuracy: 0.9955 - val_loss: 0.3731 - val_accuracy: 0.8885\n",
            "Epoch 47/75\n",
            "6/6 [==============================] - 11s 2s/step - loss: 0.0301 - accuracy: 0.9970 - val_loss: 0.3910 - val_accuracy: 0.8808\n",
            "Epoch 48/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0293 - accuracy: 0.9985 - val_loss: 0.3675 - val_accuracy: 0.8885\n",
            "Epoch 49/75\n",
            "6/6 [==============================] - 14s 2s/step - loss: 0.0320 - accuracy: 1.0000 - val_loss: 0.3930 - val_accuracy: 0.8808\n",
            "Epoch 50/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0286 - accuracy: 0.9955 - val_loss: 0.4043 - val_accuracy: 0.8769\n",
            "Epoch 51/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0304 - accuracy: 0.9970 - val_loss: 0.4107 - val_accuracy: 0.8731\n",
            "Epoch 52/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0343 - accuracy: 0.9985 - val_loss: 0.3636 - val_accuracy: 0.8885\n",
            "Epoch 53/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0338 - accuracy: 0.9970 - val_loss: 0.4214 - val_accuracy: 0.8885\n",
            "Epoch 54/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0318 - accuracy: 0.9970 - val_loss: 0.3795 - val_accuracy: 0.8885\n",
            "Epoch 55/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0248 - accuracy: 0.9985 - val_loss: 0.4043 - val_accuracy: 0.8846\n",
            "Epoch 56/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0232 - accuracy: 1.0000 - val_loss: 0.3743 - val_accuracy: 0.9000\n",
            "Epoch 57/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0228 - accuracy: 1.0000 - val_loss: 0.3713 - val_accuracy: 0.8923\n",
            "Epoch 58/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0219 - accuracy: 1.0000 - val_loss: 0.4174 - val_accuracy: 0.8769\n",
            "Epoch 59/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0216 - accuracy: 1.0000 - val_loss: 0.3769 - val_accuracy: 0.8846\n",
            "Epoch 60/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0193 - accuracy: 1.0000 - val_loss: 0.3780 - val_accuracy: 0.9038\n",
            "Epoch 61/75\n",
            "6/6 [==============================] - 11s 2s/step - loss: 0.0172 - accuracy: 1.0000 - val_loss: 0.4024 - val_accuracy: 0.8885\n",
            "Epoch 62/75\n",
            "6/6 [==============================] - 14s 2s/step - loss: 0.0180 - accuracy: 1.0000 - val_loss: 0.3906 - val_accuracy: 0.8885\n",
            "Epoch 63/75\n",
            "6/6 [==============================] - 11s 2s/step - loss: 0.0166 - accuracy: 1.0000 - val_loss: 0.3813 - val_accuracy: 0.8962\n",
            "Epoch 64/75\n",
            "6/6 [==============================] - 11s 2s/step - loss: 0.0152 - accuracy: 1.0000 - val_loss: 0.3833 - val_accuracy: 0.8846\n",
            "Epoch 65/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0140 - accuracy: 1.0000 - val_loss: 0.3866 - val_accuracy: 0.8808\n",
            "Epoch 66/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0138 - accuracy: 1.0000 - val_loss: 0.3820 - val_accuracy: 0.8923\n",
            "Epoch 67/75\n",
            "6/6 [==============================] - 14s 2s/step - loss: 0.0138 - accuracy: 1.0000 - val_loss: 0.3875 - val_accuracy: 0.8769\n",
            "Epoch 68/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0142 - accuracy: 1.0000 - val_loss: 0.4004 - val_accuracy: 0.8885\n",
            "Epoch 69/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0129 - accuracy: 1.0000 - val_loss: 0.3872 - val_accuracy: 0.9000\n",
            "Epoch 70/75\n",
            "6/6 [==============================] - 14s 2s/step - loss: 0.0129 - accuracy: 1.0000 - val_loss: 0.4035 - val_accuracy: 0.8885\n",
            "Epoch 71/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0129 - accuracy: 1.0000 - val_loss: 0.3914 - val_accuracy: 0.9000\n",
            "Epoch 72/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0119 - accuracy: 1.0000 - val_loss: 0.3964 - val_accuracy: 0.8808\n",
            "Epoch 73/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0114 - accuracy: 1.0000 - val_loss: 0.4007 - val_accuracy: 0.8808\n",
            "Epoch 74/75\n",
            "6/6 [==============================] - 12s 2s/step - loss: 0.0114 - accuracy: 1.0000 - val_loss: 0.3984 - val_accuracy: 0.8846\n",
            "Epoch 75/75\n",
            "6/6 [==============================] - 13s 2s/step - loss: 0.0108 - accuracy: 1.0000 - val_loss: 0.3954 - val_accuracy: 0.8885\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nD8untNHd-4L",
        "outputId": "ea776b99-fdf1-43fd-ef0c-140f609e0d19",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "min(history.history['val_loss'])"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.33798694610595703"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FeJ3EnARdiq8",
        "outputId": "aea3b9c1-4549-424f-e78e-e184845f0ade",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        }
      },
      "source": [
        "plot_loss(history)"
      ],
      "execution_count": 11,
      "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"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pJOatplUdnhF",
        "outputId": "e4998001-0097-4976-94a7-eca317cfa3ed",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        }
      },
      "source": [
        "plot_accuracy(history)"
      ],
      "execution_count": 9,
      "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"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "v41FLwCGebFc"
      },
      "source": [
        "Epoch 28/75\n",
        "6/6 [==============================] - 14s 2s/step - loss: 0.1101 - accuracy: 0.9653 - val_loss: 0.3380 - val_accuracy: 0.8923"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R7U-GjRUfLTh"
      },
      "source": [
        "# **Best CNN**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8pU82B52kiSi"
      },
      "source": [
        "train_datagen = ImageDataGenerator(rescale=1.0/255.0,horizontal_flip=True,zoom_range=0.1,shear_range=0.1)\n",
        "test_datagen = ImageDataGenerator(rescale=1.0/255.0)"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i2PWokbG7S8F",
        "outputId": "7f1470f1-9a36-4104-a1dd-894eec686fd5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "train_augmented_iterator = train_datagen.flow_from_directory(os.path.join(dataset_dir, 'train'), class_mode='binary', batch_size=128, target_size=(200, 200))\n",
        "test_augmented_iterator = test_datagen.flow_from_directory(os.path.join(dataset_dir, 'test'), class_mode='binary', batch_size=128, target_size=(200, 200))"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 662 images belonging to 2 classes.\n",
            "Found 260 images belonging to 2 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wxLDfa8iiUyJ",
        "outputId": "7755ad97-772d-4a80-b900-ff98c6fb198b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 765
        }
      },
      "source": [
        "model = Sequential()\n",
        "\n",
        "model.add(Conv2D(32, (3, 3), padding='same', input_shape=(200,200,3), activation='relu'))\n",
        "model.add(Conv2D(32, (3, 3), padding='same', activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(64, (3, 3), padding='same', activation='relu'))\n",
        "model.add(Conv2D(64, (3, 3), padding='same', activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(128, (3, 3), padding='same', activation='relu'))\n",
        "model.add(Conv2D(128, (3, 3), padding='same', activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(256, (3, 3), padding='same', activation='relu'))\n",
        "model.add(Conv2D(256, (3, 3), padding='same', activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Flatten())\n",
        "model.add(Dense(256, activation='relu'))\n",
        "model.add(Dropout(0.5))\n",
        "\n",
        "model.add(Dense(256, activation='relu'))\n",
        "model.add(Dropout(0.5))\n",
        "\n",
        "model.add(Dense(1,activation='sigmoid'))\n",
        "\n",
        "opt=RMSprop(lr=0.0001)\n",
        "\n",
        "model.compile(loss='binary_crossentropy',optimizer=opt,metrics=['accuracy'])\n",
        "\n",
        "model.summary()"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_3\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_17 (Conv2D)           (None, 200, 200, 32)      896       \n",
            "_________________________________________________________________\n",
            "conv2d_18 (Conv2D)           (None, 200, 200, 32)      9248      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_9 (MaxPooling2 (None, 100, 100, 32)      0         \n",
            "_________________________________________________________________\n",
            "conv2d_19 (Conv2D)           (None, 100, 100, 64)      18496     \n",
            "_________________________________________________________________\n",
            "conv2d_20 (Conv2D)           (None, 100, 100, 64)      36928     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_10 (MaxPooling (None, 50, 50, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_21 (Conv2D)           (None, 50, 50, 128)       73856     \n",
            "_________________________________________________________________\n",
            "conv2d_22 (Conv2D)           (None, 50, 50, 128)       147584    \n",
            "_________________________________________________________________\n",
            "max_pooling2d_11 (MaxPooling (None, 25, 25, 128)       0         \n",
            "_________________________________________________________________\n",
            "conv2d_23 (Conv2D)           (None, 25, 25, 256)       295168    \n",
            "_________________________________________________________________\n",
            "conv2d_24 (Conv2D)           (None, 25, 25, 256)       590080    \n",
            "_________________________________________________________________\n",
            "max_pooling2d_12 (MaxPooling (None, 12, 12, 256)       0         \n",
            "_________________________________________________________________\n",
            "flatten_3 (Flatten)          (None, 36864)             0         \n",
            "_________________________________________________________________\n",
            "dense_6 (Dense)              (None, 256)               9437440   \n",
            "_________________________________________________________________\n",
            "dropout_2 (Dropout)          (None, 256)               0         \n",
            "_________________________________________________________________\n",
            "dense_7 (Dense)              (None, 256)               65792     \n",
            "_________________________________________________________________\n",
            "dropout_3 (Dropout)          (None, 256)               0         \n",
            "_________________________________________________________________\n",
            "dense_8 (Dense)              (None, 1)                 257       \n",
            "=================================================================\n",
            "Total params: 10,675,745\n",
            "Trainable params: 10,675,745\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aEbwT2zTsBwO",
        "outputId": "87339b07-c631-4fc4-9a71-f1f5ce118714",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "history=model.fit(train_augmented_iterator, steps_per_epoch=len(train_augmented_iterator), validation_data=test_augmented_iterator, validation_steps=len(test_augmented_iterator), epochs=75)"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/75\n",
            "6/6 [==============================] - 18s 3s/step - loss: 0.6885 - accuracy: 0.5060 - val_loss: 0.6657 - val_accuracy: 0.5000\n",
            "Epoch 2/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.6385 - accuracy: 0.5151 - val_loss: 0.6192 - val_accuracy: 0.8077\n",
            "Epoch 3/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.5709 - accuracy: 0.7417 - val_loss: 0.5321 - val_accuracy: 0.8423\n",
            "Epoch 4/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.5490 - accuracy: 0.8082 - val_loss: 0.4998 - val_accuracy: 0.8462\n",
            "Epoch 5/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.4630 - accuracy: 0.8384 - val_loss: 0.6593 - val_accuracy: 0.7692\n",
            "Epoch 6/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.4358 - accuracy: 0.8776 - val_loss: 0.4006 - val_accuracy: 0.8615\n",
            "Epoch 7/75\n",
            "6/6 [==============================] - 18s 3s/step - loss: 0.4306 - accuracy: 0.8489 - val_loss: 0.4084 - val_accuracy: 0.8577\n",
            "Epoch 8/75\n",
            "6/6 [==============================] - 19s 3s/step - loss: 0.3712 - accuracy: 0.8807 - val_loss: 0.6708 - val_accuracy: 0.7154\n",
            "Epoch 9/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.4012 - accuracy: 0.8489 - val_loss: 0.3866 - val_accuracy: 0.8692\n",
            "Epoch 10/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.3148 - accuracy: 0.8988 - val_loss: 0.5735 - val_accuracy: 0.7692\n",
            "Epoch 11/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.4050 - accuracy: 0.8474 - val_loss: 0.3566 - val_accuracy: 0.8769\n",
            "Epoch 12/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.2839 - accuracy: 0.9124 - val_loss: 0.3360 - val_accuracy: 0.8885\n",
            "Epoch 13/75\n",
            "6/6 [==============================] - 20s 3s/step - loss: 0.2751 - accuracy: 0.9154 - val_loss: 0.5734 - val_accuracy: 0.7923\n",
            "Epoch 14/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.3304 - accuracy: 0.8927 - val_loss: 0.3141 - val_accuracy: 0.9000\n",
            "Epoch 15/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.2895 - accuracy: 0.8822 - val_loss: 0.3194 - val_accuracy: 0.8846\n",
            "Epoch 16/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.2594 - accuracy: 0.9048 - val_loss: 0.5710 - val_accuracy: 0.7577\n",
            "Epoch 17/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.3304 - accuracy: 0.8837 - val_loss: 0.4063 - val_accuracy: 0.8538\n",
            "Epoch 18/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.2836 - accuracy: 0.9033 - val_loss: 0.3584 - val_accuracy: 0.8577\n",
            "Epoch 19/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.2341 - accuracy: 0.9260 - val_loss: 0.2667 - val_accuracy: 0.9038\n",
            "Epoch 20/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.3112 - accuracy: 0.8761 - val_loss: 0.2752 - val_accuracy: 0.9115\n",
            "Epoch 21/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.2721 - accuracy: 0.8776 - val_loss: 0.2595 - val_accuracy: 0.9115\n",
            "Epoch 22/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.2062 - accuracy: 0.9230 - val_loss: 0.2898 - val_accuracy: 0.8962\n",
            "Epoch 23/75\n",
            "6/6 [==============================] - 19s 3s/step - loss: 0.2319 - accuracy: 0.9169 - val_loss: 0.2380 - val_accuracy: 0.9115\n",
            "Epoch 24/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.2143 - accuracy: 0.9245 - val_loss: 0.3057 - val_accuracy: 0.9000\n",
            "Epoch 25/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.1799 - accuracy: 0.9426 - val_loss: 0.2702 - val_accuracy: 0.8962\n",
            "Epoch 26/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.4264 - accuracy: 0.8429 - val_loss: 0.2501 - val_accuracy: 0.9077\n",
            "Epoch 27/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1686 - accuracy: 0.9471 - val_loss: 0.2218 - val_accuracy: 0.9269\n",
            "Epoch 28/75\n",
            "6/6 [==============================] - 19s 3s/step - loss: 0.2162 - accuracy: 0.9033 - val_loss: 0.2231 - val_accuracy: 0.9269\n",
            "Epoch 29/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.1653 - accuracy: 0.9366 - val_loss: 0.2224 - val_accuracy: 0.9308\n",
            "Epoch 30/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.3242 - accuracy: 0.8610 - val_loss: 0.2278 - val_accuracy: 0.9231\n",
            "Epoch 31/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.1537 - accuracy: 0.9547 - val_loss: 0.2210 - val_accuracy: 0.9308\n",
            "Epoch 32/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1516 - accuracy: 0.9471 - val_loss: 0.3189 - val_accuracy: 0.8846\n",
            "Epoch 33/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1888 - accuracy: 0.9290 - val_loss: 0.3587 - val_accuracy: 0.8808\n",
            "Epoch 34/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1647 - accuracy: 0.9426 - val_loss: 0.4024 - val_accuracy: 0.8154\n",
            "Epoch 35/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.2484 - accuracy: 0.8973 - val_loss: 0.2094 - val_accuracy: 0.9385\n",
            "Epoch 36/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1686 - accuracy: 0.9335 - val_loss: 0.2451 - val_accuracy: 0.9154\n",
            "Epoch 37/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1979 - accuracy: 0.9381 - val_loss: 0.2882 - val_accuracy: 0.8962\n",
            "Epoch 38/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1365 - accuracy: 0.9502 - val_loss: 0.2821 - val_accuracy: 0.9038\n",
            "Epoch 39/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1906 - accuracy: 0.9230 - val_loss: 0.2079 - val_accuracy: 0.9385\n",
            "Epoch 40/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1287 - accuracy: 0.9562 - val_loss: 0.2047 - val_accuracy: 0.9423\n",
            "Epoch 41/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1918 - accuracy: 0.9215 - val_loss: 0.2174 - val_accuracy: 0.9308\n",
            "Epoch 42/75\n",
            "6/6 [==============================] - 19s 3s/step - loss: 0.1759 - accuracy: 0.9396 - val_loss: 0.2100 - val_accuracy: 0.9308\n",
            "Epoch 43/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.2179 - accuracy: 0.9215 - val_loss: 0.2146 - val_accuracy: 0.9346\n",
            "Epoch 44/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1262 - accuracy: 0.9517 - val_loss: 0.3838 - val_accuracy: 0.8500\n",
            "Epoch 45/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1825 - accuracy: 0.9275 - val_loss: 0.3844 - val_accuracy: 0.8808\n",
            "Epoch 46/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1842 - accuracy: 0.9260 - val_loss: 0.1965 - val_accuracy: 0.9423\n",
            "Epoch 47/75\n",
            "6/6 [==============================] - 19s 3s/step - loss: 0.1291 - accuracy: 0.9517 - val_loss: 0.2433 - val_accuracy: 0.9154\n",
            "Epoch 48/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1983 - accuracy: 0.9335 - val_loss: 0.1953 - val_accuracy: 0.9423\n",
            "Epoch 49/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1243 - accuracy: 0.9607 - val_loss: 0.2051 - val_accuracy: 0.9385\n",
            "Epoch 50/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1189 - accuracy: 0.9577 - val_loss: 0.2112 - val_accuracy: 0.9385\n",
            "Epoch 51/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.1040 - accuracy: 0.9668 - val_loss: 0.2620 - val_accuracy: 0.9000\n",
            "Epoch 52/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.2221 - accuracy: 0.9184 - val_loss: 0.2054 - val_accuracy: 0.9423\n",
            "Epoch 53/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.1218 - accuracy: 0.9592 - val_loss: 0.2108 - val_accuracy: 0.9385\n",
            "Epoch 54/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.0913 - accuracy: 0.9728 - val_loss: 0.4047 - val_accuracy: 0.8577\n",
            "Epoch 55/75\n",
            "6/6 [==============================] - 19s 3s/step - loss: 0.3033 - accuracy: 0.8852 - val_loss: 0.2112 - val_accuracy: 0.9269\n",
            "Epoch 56/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1130 - accuracy: 0.9532 - val_loss: 0.2113 - val_accuracy: 0.9385\n",
            "Epoch 57/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.0933 - accuracy: 0.9728 - val_loss: 0.2213 - val_accuracy: 0.9231\n",
            "Epoch 58/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.1299 - accuracy: 0.9426 - val_loss: 0.2059 - val_accuracy: 0.9423\n",
            "Epoch 59/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1744 - accuracy: 0.9320 - val_loss: 0.3489 - val_accuracy: 0.8577\n",
            "Epoch 60/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1110 - accuracy: 0.9607 - val_loss: 0.1989 - val_accuracy: 0.9423\n",
            "Epoch 61/75\n",
            "6/6 [==============================] - 19s 3s/step - loss: 0.0865 - accuracy: 0.9698 - val_loss: 0.2086 - val_accuracy: 0.9500\n",
            "Epoch 62/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1265 - accuracy: 0.9502 - val_loss: 0.2105 - val_accuracy: 0.9269\n",
            "Epoch 63/75\n",
            "6/6 [==============================] - 19s 3s/step - loss: 0.0880 - accuracy: 0.9698 - val_loss: 0.2148 - val_accuracy: 0.9385\n",
            "Epoch 64/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1539 - accuracy: 0.9411 - val_loss: 0.2771 - val_accuracy: 0.9154\n",
            "Epoch 65/75\n",
            "6/6 [==============================] - 19s 3s/step - loss: 0.1576 - accuracy: 0.9381 - val_loss: 0.2160 - val_accuracy: 0.9269\n",
            "Epoch 66/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.0791 - accuracy: 0.9728 - val_loss: 0.2907 - val_accuracy: 0.9154\n",
            "Epoch 67/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.1619 - accuracy: 0.9305 - val_loss: 0.4029 - val_accuracy: 0.8308\n",
            "Epoch 68/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1306 - accuracy: 0.9502 - val_loss: 0.1921 - val_accuracy: 0.9500\n",
            "Epoch 69/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.0689 - accuracy: 0.9743 - val_loss: 0.2174 - val_accuracy: 0.9423\n",
            "Epoch 70/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.2272 - accuracy: 0.9260 - val_loss: 0.2695 - val_accuracy: 0.8962\n",
            "Epoch 71/75\n",
            "6/6 [==============================] - 19s 3s/step - loss: 0.0840 - accuracy: 0.9698 - val_loss: 0.1953 - val_accuracy: 0.9500\n",
            "Epoch 72/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1080 - accuracy: 0.9562 - val_loss: 0.2358 - val_accuracy: 0.9231\n",
            "Epoch 73/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.0765 - accuracy: 0.9758 - val_loss: 0.2262 - val_accuracy: 0.9423\n",
            "Epoch 74/75\n",
            "6/6 [==============================] - 17s 3s/step - loss: 0.1642 - accuracy: 0.9335 - val_loss: 0.2033 - val_accuracy: 0.9385\n",
            "Epoch 75/75\n",
            "6/6 [==============================] - 16s 3s/step - loss: 0.0646 - accuracy: 0.9789 - val_loss: 0.1991 - val_accuracy: 0.9462\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eNM2o4KKlyb2",
        "outputId": "f6915ff3-3b47-4d69-92ff-3da7f23241fd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "min(history.history['val_loss'])"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.19208741188049316"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZVDHpoev96Zr",
        "outputId": "85c68a2c-738f-4fc1-ff91-9123cc9546b1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        }
      },
      "source": [
        "plot_loss(history)"
      ],
      "execution_count": 22,
      "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"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UsInz2-j9_eA",
        "outputId": "100081cf-f327-478b-a9a1-6d0130593e11",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        }
      },
      "source": [
        "plot_accuracy(history)"
      ],
      "execution_count": 23,
      "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"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YRKCWBnjl8ui"
      },
      "source": [
        "Epoch 71/75\n",
        "6/6 [==============================] - 19s 3s/step - loss: 0.0840 - accuracy: 0.9698 - val_loss: 0.1953 - val_accuracy: 0.9500"
      ]
    }
  ]
}