ml/ML/cnn/lab1/lab1_dogscats.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "_epZXfryWUPb"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd \n",
        "from keras.preprocessing.image import ImageDataGenerator, load_img\n",
        "\n",
        "import os\n",
        "import random\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# initialize parameters\n",
        "EPOCHS = 50\n",
        "IMAGE_SIZE = (128, 128)\n",
        "CHANNELS = 3\n",
        "TRAIN_FOLDER = 'train'\n",
        "TEST_FOLDER = 'test1'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Kt1NKfCnWUPi",
        "outputId": "649543ec-cb09-4ae0-d716-1d292720abc9"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "            filename label\n",
            "13533  dog.10927.jpg   dog\n",
            "20489   dog.5939.jpg   dog\n",
            "19649   dog.5182.jpg   dog\n",
            "24719   dog.9746.jpg   dog\n",
            "1380   cat.11239.jpg   cat\n"
          ]
        }
      ],
      "source": [
        "filenames = os.listdir(TRAIN_FOLDER)\n",
        "\n",
        "train_set = {}\n",
        "\n",
        "# load images and labels\n",
        "\n",
        "for file in filenames:\n",
        "    if file[:3] == 'cat':\n",
        "        train_set[file] = 'cat'\n",
        "    else:\n",
        "        train_set[file] = 'dog'\n",
        "\n",
        "train_set = pd.DataFrame(train_set.items(), columns=['filename', 'label'])\n",
        "print(train_set.sample(frac=1).head())\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FOpA2L0EWUPk",
        "outputId": "2333bf33-c963-4af1-ecc2-e3efb3ed4e76"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d (Conv2D)             (None, 126, 126, 32)      896       \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 63, 63, 32)       0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 61, 61, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 30, 30, 64)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_2 (Conv2D)           (None, 28, 28, 128)       73856     \n",
            "                                                                 \n",
            " max_pooling2d_2 (MaxPooling  (None, 14, 14, 128)      0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 25088)             0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 512)               12845568  \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 2)                 1026      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 12,939,842\n",
            "Trainable params: 12,939,842\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Conv2D, MaxPooling2D, Dropout, Flatten, Dense, Activation, BatchNormalization\n",
        "\n",
        "# Build CNN\n",
        "\n",
        "cnn = Sequential()\n",
        "\n",
        "# Convolution + Pooling\n",
        "\n",
        "cnn.add(Conv2D(32, (3,3), activation='relu', input_shape=(IMAGE_SIZE[0], IMAGE_SIZE[1], CHANNELS)))\n",
        "cnn.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "cnn.add(Conv2D(64, (3,3), activation='relu'))\n",
        "cnn.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "cnn.add(Conv2D(128, (3,3), activation='relu'))\n",
        "cnn.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "# Flattening\n",
        "\n",
        "cnn.add(Flatten())\n",
        "\n",
        "# NN\n",
        "\n",
        "cnn.add(Dense(512, activation='relu'))\n",
        "cnn.add(Dense(2, activation='softmax'))\n",
        "\n",
        "cnn.compile(loss='categorical_crossentropy', optimizer='rmsprop', metrics=['accuracy'])\n",
        "\n",
        "cnn.summary()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "anB0xTQqZ2xq"
      },
      "outputs": [],
      "source": [
        "from keras.callbacks import EarlyStopping\n",
        "\n",
        "# turn on early stopping\n",
        "earlystop = EarlyStopping(patience=10)\n",
        "callbacks = [earlystop]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "7ee7SA0EaeZb"
      },
      "outputs": [],
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# split data into train and test sets\n",
        "train, validate = train_test_split(train_set, test_size=0.20, random_state=42)\n",
        "train = train.reset_index(drop=True)\n",
        "validate = validate.reset_index(drop=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xnqb08HQbIQQ",
        "outputId": "43d33371-d019-4183-efc0-c7c63b32a63d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Found 20000 validated image filenames belonging to 2 classes.\n",
            "Found 5000 validated image filenames belonging to 2 classes.\n"
          ]
        }
      ],
      "source": [
        "# create generators to transform and load images\n",
        "rescale = ImageDataGenerator(\n",
        "    rescale=1./255,\n",
        "    rotation_range=15,\n",
        "    shear_range=0.1,\n",
        "    zoom_range=0.2,\n",
        "    horizontal_flip=True,\n",
        "    width_shift_range=0.1,\n",
        "    height_shift_range=0.1\n",
        ")\n",
        "\n",
        "generator = rescale.flow_from_dataframe(\n",
        "    train, \n",
        "    \"train/\", \n",
        "    x_col='filename',\n",
        "    y_col='label',\n",
        "    target_size=IMAGE_SIZE,\n",
        "    class_mode='categorical',\n",
        "    batch_size=16\n",
        ")\n",
        "\n",
        "validation_datagen = ImageDataGenerator(rescale=1./255)\n",
        "validation_generator = validation_datagen.flow_from_dataframe(\n",
        "    validate, \n",
        "    \"train\", \n",
        "    x_col='filename',\n",
        "    y_col='label',\n",
        "    target_size=IMAGE_SIZE,\n",
        "    class_mode='categorical',\n",
        "    batch_size=16\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/"
        },
        "id": "2m2lOT_icYV4",
        "outputId": "d52a1011-e20e-46f0-8f7d-941b4f63e7ee"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/50\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "C:\\Users\\rushi\\AppData\\Local\\Temp\\ipykernel_35232\\1210287814.py:5: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
            "  history = cnn.fit_generator(\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1250/1250 [==============================] - 300s 239ms/step - loss: 0.6365 - accuracy: 0.6486 - val_loss: 0.5296 - val_accuracy: 0.7374\n",
            "Epoch 2/50\n",
            "1250/1250 [==============================] - 289s 231ms/step - loss: 0.5468 - accuracy: 0.7304 - val_loss: 0.4601 - val_accuracy: 0.7794\n",
            "Epoch 3/50\n",
            "1250/1250 [==============================] - 290s 232ms/step - loss: 0.5125 - accuracy: 0.7549 - val_loss: 0.4411 - val_accuracy: 0.8007\n",
            "Epoch 4/50\n",
            "1250/1250 [==============================] - 290s 232ms/step - loss: 0.4893 - accuracy: 0.7722 - val_loss: 0.4371 - val_accuracy: 0.8055\n",
            "Epoch 5/50\n",
            "1250/1250 [==============================] - 277s 221ms/step - loss: 0.4721 - accuracy: 0.7836 - val_loss: 0.5153 - val_accuracy: 0.7348\n",
            "Epoch 6/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.4611 - accuracy: 0.7901 - val_loss: 0.3957 - val_accuracy: 0.8145\n",
            "Epoch 7/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.4496 - accuracy: 0.8013 - val_loss: 0.6899 - val_accuracy: 0.7053\n",
            "Epoch 8/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.4390 - accuracy: 0.8061 - val_loss: 0.4314 - val_accuracy: 0.8215\n",
            "Epoch 9/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.4280 - accuracy: 0.8130 - val_loss: 0.3720 - val_accuracy: 0.8287\n",
            "Epoch 10/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.4137 - accuracy: 0.8177 - val_loss: 0.3544 - val_accuracy: 0.8446\n",
            "Epoch 11/50\n",
            "1250/1250 [==============================] - 276s 221ms/step - loss: 0.4170 - accuracy: 0.8215 - val_loss: 0.3334 - val_accuracy: 0.8568\n",
            "Epoch 12/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.4040 - accuracy: 0.8257 - val_loss: 0.3409 - val_accuracy: 0.8588\n",
            "Epoch 13/50\n",
            "1250/1250 [==============================] - 277s 221ms/step - loss: 0.4041 - accuracy: 0.8274 - val_loss: 0.4161 - val_accuracy: 0.8604\n",
            "Epoch 14/50\n",
            "1250/1250 [==============================] - 278s 222ms/step - loss: 0.4098 - accuracy: 0.8268 - val_loss: 0.3689 - val_accuracy: 0.8472\n",
            "Epoch 15/50\n",
            "1250/1250 [==============================] - 277s 221ms/step - loss: 0.4004 - accuracy: 0.8301 - val_loss: 0.3040 - val_accuracy: 0.8738\n",
            "Epoch 16/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.3911 - accuracy: 0.8332 - val_loss: 0.4173 - val_accuracy: 0.8405\n",
            "Epoch 17/50\n",
            "1250/1250 [==============================] - 278s 222ms/step - loss: 0.3914 - accuracy: 0.8339 - val_loss: 0.4132 - val_accuracy: 0.8237\n",
            "Epoch 18/50\n",
            "1250/1250 [==============================] - 278s 222ms/step - loss: 0.3910 - accuracy: 0.8351 - val_loss: 0.3672 - val_accuracy: 0.8492\n",
            "Epoch 19/50\n",
            "1250/1250 [==============================] - 277s 221ms/step - loss: 0.3840 - accuracy: 0.8349 - val_loss: 0.3191 - val_accuracy: 0.8588\n",
            "Epoch 20/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.3833 - accuracy: 0.8415 - val_loss: 0.4091 - val_accuracy: 0.8309\n",
            "Epoch 21/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.3782 - accuracy: 0.8419 - val_loss: 0.3698 - val_accuracy: 0.8468\n",
            "Epoch 22/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.3778 - accuracy: 0.8418 - val_loss: 0.3507 - val_accuracy: 0.8710\n",
            "Epoch 23/50\n",
            "1250/1250 [==============================] - 277s 222ms/step - loss: 0.3793 - accuracy: 0.8422 - val_loss: 0.3374 - val_accuracy: 0.8640\n",
            "Epoch 24/50\n",
            "1250/1250 [==============================] - 278s 222ms/step - loss: 0.3758 - accuracy: 0.8417 - val_loss: 0.3118 - val_accuracy: 0.8658\n",
            "Epoch 25/50\n",
            "1250/1250 [==============================] - 278s 222ms/step - loss: 0.3635 - accuracy: 0.8486 - val_loss: 0.3619 - val_accuracy: 0.8578\n"
          ]
        }
      ],
      "source": [
        "total_train = train.shape[0]\n",
        "total_validate = validate.shape[0]\n",
        "batch_size = 16\n",
        "\n",
        "# generate cnn and train\n",
        "\n",
        "history = cnn.fit_generator(\n",
        "    generator, \n",
        "    epochs=EPOCHS,\n",
        "    validation_data=validation_generator,\n",
        "    validation_steps=total_validate//batch_size,\n",
        "    steps_per_epoch=total_train//batch_size,\n",
        "    callbacks=callbacks\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "MNj-kUEMD6t5"
      },
      "outputs": [],
      "source": [
        "\n",
        "# save model and weights\n",
        "cnn.save(\"cnn.h5\")\n",
        "cnn.save_weights(\"weights.h5\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "id": "gLC08WZ8YjKC"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x864 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# plot data\n",
        "fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(12, 12))\n",
        "ax1.plot(history.history['loss'], color='b', label=\"Training loss\")\n",
        "ax1.plot(history.history['val_loss'], color='r', label=\"Validation loss\")\n",
        "ax1.set_xticks(np.arange(1, 25, 1))\n",
        "ax1.legend(loc='best', shadow=True)\n",
        "\n",
        "ax2.plot(history.history['accuracy'], color='b', label=\"Training accuracy\")\n",
        "ax2.plot(history.history['val_accuracy'], color='r',label=\"Validation accuracy\")\n",
        "ax2.set_xticks(np.arange(1, 25, 1))\n",
        "ax2.legend(loc='best', shadow=True)\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [],
      "name": "lab1.ipynb",
      "provenance": []
    },
    "interpreter": {
      "hash": "2db524e06e9f5f4ffedc911c917cb75e12dbc923643829bf417064a77eb14d37"
    },
    "kernelspec": {
      "display_name": "Python 3.8.5 64-bit",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.6"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}