CAT_DOG_TRAIN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "CAT_DOG_TRAIN.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/swamivisal/Pet-Image-Classification/blob/main/CAT_DOG_TRAIN.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gOn45gCu_adw"
      },
      "source": [
        "##**Importing the libraries**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rR7S6cCf-Fk4"
      },
      "source": [
        "import numpy as np\n",
        "import cv2\n",
        "from google.colab.patches import cv2_imshow\n",
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense,Activation,Flatten,Conv2D,MaxPool2D,Dropout\n",
        "from random import shuffle\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1V9pHZHAAOCg"
      },
      "source": [
        "##**Preprocessing the training and validation data**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 735
        },
        "id": "oVcxDmQV-IqC",
        "outputId": "3e2e9e5b-d215-4cce-af04-a0412744c6a3"
      },
      "source": [
        "\n",
        "#def augment(img):\n",
        "#M=cv2.getRotationMatrix2D((25,25),np.random.randint(-10,11),1)\n",
        "#img=cv2.warpAffine(img,M,(50,50))\n",
        "#return img\n",
        "\n",
        "\n",
        "path = \"/content/drive/MyDrive/Computer Vision/TRAIN.npy\"\n",
        "\n",
        "classes = [\"Cat\",\"Dog\"]\n",
        "dataset = np.load(path,allow_pickle=True)\n",
        "print(dataset.shape)\n",
        "shuffle(dataset)\n",
        "inputs= [] \n",
        "targets = []\n",
        "\n",
        "for x,y in dataset[:20000]:\n",
        "  inputs.append(x)\n",
        "  targets.append(y)\n",
        "\n",
        "inputs=np.array(inputs)\n",
        "targets=np.array(targets)\n",
        "print(inputs.shape)\n",
        "print(targets.shape)\n",
        "normalised_inputs=inputs/255.0  \n",
        "for i in range(10):\n",
        "  cv2_imshow(inputs[i])\n",
        "  print(classes[targets[i]])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(24946, 2)\n",
            "(20000, 50, 50, 3)\n",
            "(20000,)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=50x50 at 0x7F960974C0D0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Cat\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=50x50 at 0x7F9607A22B50>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Cat\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=50x50 at 0x7F9607A229D0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Cat\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=50x50 at 0x7F9607A22A90>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Cat\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=50x50 at 0x7F9607A22A50>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Dog\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=50x50 at 0x7F9607A22C50>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Cat\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=50x50 at 0x7F9607A22A90>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Cat\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=50x50 at 0x7F9609BCBF50>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Cat\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=50x50 at 0x7F9607A22C90>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Dog\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=50x50 at 0x7F9609BCB910>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Dog\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rAorrjgcBBRk"
      },
      "source": [
        "##**Creating the Model Architecture**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CI_7q-Z7Ap5v"
      },
      "source": [
        "model = keras.Sequential()\n",
        "model.add(Conv2D(32,(3,3),padding=\"same\",input_shape=normalised_inputs.shape[1:]))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(0.3))\n",
        "model.add(MaxPool2D(pool_size=(2,2),strides=(2,2)))\n",
        "model.add(Conv2D(64,(3,3),padding=\"same\"))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(0.3))\n",
        "model.add(MaxPool2D(pool_size=(2,2),strides=(2,2)))\n",
        "model.add(Conv2D(128,(3,3),padding=\"same\"))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(0.3))\n",
        "model.add(MaxPool2D(pool_size=(2,2),strides=(2,2)))\n",
        "model.add(Conv2D(256,(3,3),padding=\"same\"))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(0.3))\n",
        "model.add(MaxPool2D(pool_size=(2,2),strides=(2,2)))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(128))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(0.3))\n",
        "model.add(Dense(1))\n",
        "model.add(Activation('sigmoid'))\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gr4wL0peApAl"
      },
      "source": [
        "##**Compiling the CNN and training the model on training data and saving the best model with the less validation loss**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tWezUazJ6TUQ",
        "outputId": "c10de4c1-e281-4041-bf10-d844b2fb32e5"
      },
      "source": [
        "model.compile(loss=\"binary_crossentropy\",optimizer=keras.optimizers.Adam(learning_rate=0.001),metrics=[\"accuracy\"])\n",
        "filepath=\"/content/drive/MyDrive/Computer Vision/Model/best.hdf5\"\n",
        "checkpoint=ModelCheckpoint(filepath,monitor='val_loss',verbose=1,save_best_only=True,mode='min')\n",
        "model.fit(normalised_inputs,targets,validation_split=0.05,batch_size=64,epochs=20,callbacks=[checkpoint])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "297/297 [==============================] - 7s 22ms/step - loss: 0.6424 - accuracy: 0.6185 - val_loss: 0.6135 - val_accuracy: 0.6860\n",
            "\n",
            "Epoch 00001: val_loss improved from inf to 0.61346, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 2/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.5171 - accuracy: 0.7453 - val_loss: 0.5185 - val_accuracy: 0.7690\n",
            "\n",
            "Epoch 00002: val_loss improved from 0.61346 to 0.51849, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 3/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.4231 - accuracy: 0.8068 - val_loss: 0.4750 - val_accuracy: 0.7920\n",
            "\n",
            "Epoch 00003: val_loss improved from 0.51849 to 0.47500, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 4/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.3541 - accuracy: 0.8455 - val_loss: 0.3997 - val_accuracy: 0.8400\n",
            "\n",
            "Epoch 00004: val_loss improved from 0.47500 to 0.39973, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 5/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.2980 - accuracy: 0.8759 - val_loss: 0.3863 - val_accuracy: 0.8250\n",
            "\n",
            "Epoch 00005: val_loss improved from 0.39973 to 0.38627, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 6/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.2460 - accuracy: 0.9004 - val_loss: 0.3543 - val_accuracy: 0.8600\n",
            "\n",
            "Epoch 00006: val_loss improved from 0.38627 to 0.35431, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 7/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.2148 - accuracy: 0.9148 - val_loss: 0.3138 - val_accuracy: 0.8670\n",
            "\n",
            "Epoch 00007: val_loss improved from 0.35431 to 0.31375, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 8/20\n",
            "297/297 [==============================] - 6s 21ms/step - loss: 0.1775 - accuracy: 0.9329 - val_loss: 0.3057 - val_accuracy: 0.8770\n",
            "\n",
            "Epoch 00008: val_loss improved from 0.31375 to 0.30567, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 9/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.1618 - accuracy: 0.9367 - val_loss: 0.3035 - val_accuracy: 0.8720\n",
            "\n",
            "Epoch 00009: val_loss improved from 0.30567 to 0.30348, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 10/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.1424 - accuracy: 0.9473 - val_loss: 0.3403 - val_accuracy: 0.8570\n",
            "\n",
            "Epoch 00010: val_loss did not improve from 0.30348\n",
            "Epoch 11/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.1234 - accuracy: 0.9524 - val_loss: 0.2775 - val_accuracy: 0.9000\n",
            "\n",
            "Epoch 00011: val_loss improved from 0.30348 to 0.27749, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 12/20\n",
            "297/297 [==============================] - 6s 21ms/step - loss: 0.1147 - accuracy: 0.9563 - val_loss: 0.2690 - val_accuracy: 0.9000\n",
            "\n",
            "Epoch 00012: val_loss improved from 0.27749 to 0.26902, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 13/20\n",
            "297/297 [==============================] - 6s 21ms/step - loss: 0.1040 - accuracy: 0.9611 - val_loss: 0.2585 - val_accuracy: 0.9080\n",
            "\n",
            "Epoch 00013: val_loss improved from 0.26902 to 0.25854, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 14/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.0977 - accuracy: 0.9645 - val_loss: 0.2524 - val_accuracy: 0.9110\n",
            "\n",
            "Epoch 00014: val_loss improved from 0.25854 to 0.25236, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 15/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.0877 - accuracy: 0.9688 - val_loss: 0.2695 - val_accuracy: 0.9030\n",
            "\n",
            "Epoch 00015: val_loss did not improve from 0.25236\n",
            "Epoch 16/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.0865 - accuracy: 0.9676 - val_loss: 0.2495 - val_accuracy: 0.9090\n",
            "\n",
            "Epoch 00016: val_loss improved from 0.25236 to 0.24951, saving model to /content/drive/MyDrive/Computer Vision/Model/best.hdf5\n",
            "Epoch 17/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.0760 - accuracy: 0.9715 - val_loss: 0.2629 - val_accuracy: 0.9030\n",
            "\n",
            "Epoch 00017: val_loss did not improve from 0.24951\n",
            "Epoch 18/20\n",
            "297/297 [==============================] - 6s 21ms/step - loss: 0.0728 - accuracy: 0.9723 - val_loss: 0.2625 - val_accuracy: 0.9010\n",
            "\n",
            "Epoch 00018: val_loss did not improve from 0.24951\n",
            "Epoch 19/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.0692 - accuracy: 0.9757 - val_loss: 0.2677 - val_accuracy: 0.8990\n",
            "\n",
            "Epoch 00019: val_loss did not improve from 0.24951\n",
            "Epoch 20/20\n",
            "297/297 [==============================] - 6s 22ms/step - loss: 0.0640 - accuracy: 0.9759 - val_loss: 0.2689 - val_accuracy: 0.9080\n",
            "\n",
            "Epoch 00020: val_loss did not improve from 0.24951\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f2725ba8b50>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    }
  ]
}