Group4_Face_mask_detection_model_EDAI.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "EDAI First achivement.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DRgg_hg-Gpp0"
      },
      "source": [
        "# Mounting the Drive"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7gsoir-8E-sr",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f8625ae5-ffd0-45d6-9194-e7c159928cc9"
      },
      "source": [
        "from google.colab import drive \n",
        "drive.mount(\"/content/drive\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XtHvXSTODtaY"
      },
      "source": [
        "# Importing essential libraries \n",
        "import tensorflow as tf\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import cv2\n",
        "from keras.preprocessing import image\n",
        "from PIL import Image\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras import layers"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-OWYYwrz5R_X"
      },
      "source": [
        "##Loading dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "57gjv8K8G8eR"
      },
      "source": [
        "# Normalizing the images using imagedatagenerator\n",
        "train_datagen = ImageDataGenerator(\n",
        "        rescale=1./255,\n",
        "        shear_range=0.1,\n",
        "        zoom_range=0.1,\n",
        "        horizontal_flip=True,\n",
        "        rotation_range=20,\n",
        "        width_shift_range = 0.1,\n",
        "        height_shift_range = 0.1,\n",
        "      )\n",
        "test_datagen = ImageDataGenerator(rescale=1./255)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gsXYsNjwclRM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "53208f34-f0f9-4799-f656-5f993903de5a"
      },
      "source": [
        "# Provided the google drive link to upload the dataset\n",
        "# Uploading images with the target size of 128 x 128 pixels in RGB format \n",
        "\n",
        "# train_generator contains the images on which the model is to be train\n",
        "train_generator = train_datagen.flow_from_directory(\n",
        "        '/content/drive/MyDrive/Others/Face Mask Dataset/Train',\n",
        "        target_size=(150,150),\n",
        "         color_mode=\"grayscale\",\n",
        "        batch_size=32,\n",
        "        class_mode='binary')\n",
        "\n",
        "# validation_generator contains the images on which we are validating the model\n",
        "validation_generator = test_datagen.flow_from_directory(\n",
        "        '/content/drive/MyDrive/Others/Face Mask Dataset/Validation',\n",
        "        target_size=(150,150),\n",
        "         color_mode=\"grayscale\",\n",
        "        batch_size=32,\n",
        "        class_mode='binary')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 10000 images belonging to 2 classes.\n",
            "Found 800 images belonging to 2 classes.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "maN1kmjxcrHE"
      },
      "source": [
        "##Library for neural network"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_OM4j-Bzcnzu"
      },
      "source": [
        "# Importing libraries which require to build model architecture\n",
        "from keras.layers import Dense\n",
        "from keras.models import Sequential\n",
        "from tensorflow.keras.layers import Flatten,BatchNormalization\n",
        "from tensorflow.keras.layers import MaxPooling2D,Conv2D\n",
        "from tensorflow.keras.layers import Input,Activation,Add\n",
        "from tensorflow.keras.models import Model \n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from tensorflow.keras.layers import Conv2D,InputLayer, Dropout, BatchNormalization, Flatten, Dense, MaxPooling2D"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ECAtmN8vR7ZD"
      },
      "source": [
        "## Creating Convolutional Neural Network Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TIKpBo97dRMr"
      },
      "source": [
        "# building seqential model\n",
        "model=Sequential()\n",
        "model.add(Conv2D(32,(3,3),input_shape=(150,150,1),activation=\"relu\"))\n",
        "model.add(MaxPooling2D((2,2)))\n",
        "model.add(BatchNormalization()) \n",
        "model.add(Conv2D(64,(3,3),activation=\"relu\"))\n",
        "model.add(MaxPooling2D((2,2)))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(128,activation=\"relu\"))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(1,activation=\"sigmoid\"))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NTlsPnAX2GOP",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f21f6b02-24e4-4434-b832-7b14bc9e43a6"
      },
      "source": [
        "# Summarizing all the layer in model\n",
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d (Conv2D)             (None, 148, 148, 32)      320       \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 74, 74, 32)       0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 74, 74, 32)       128       \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 72, 72, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 36, 36, 64)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 82944)             0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 128)               10616960  \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 128)               0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 1)                 129       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 10,636,033\n",
            "Trainable params: 10,635,969\n",
            "Non-trainable params: 64\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "s1dnegVwhxtr"
      },
      "source": [
        "model.compile(optimizer='adam',loss='binary_crossentropy',metrics=['accuracy'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "a0AUGhP2ZuYg"
      },
      "source": [
        "# Adding callback to keep a track on leaning rate and doing neccesory changes\n",
        "from keras.callbacks import EarlyStopping, ReduceLROnPlateau\n",
        "callbacks = [EarlyStopping(patience=4, monitor='val_loss', mode='min'), \n",
        "             ReduceLROnPlateau(patience=2, verbose=1)]             "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "X9pQ4f4WGLBm",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b573cca8-5ff3-4fe7-c128-764bc81f1e2d"
      },
      "source": [
        "# Training the model with 30 epochs and with the batch size of 100\n",
        "n_epochs = 30\n",
        "batch_size = 100\n",
        "history = model.fit(train_generator,  \n",
        "                           epochs=n_epochs, \n",
        "                           callbacks=callbacks,\n",
        "                           validation_data=validation_generator)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/30\n",
            "313/313 [==============================] - 71s 220ms/step - loss: 0.4974 - accuracy: 0.8510 - val_loss: 0.5629 - val_accuracy: 0.7075 - lr: 0.0010\n",
            "Epoch 2/30\n",
            "313/313 [==============================] - 66s 209ms/step - loss: 0.1833 - accuracy: 0.9296 - val_loss: 0.1180 - val_accuracy: 0.9575 - lr: 0.0010\n",
            "Epoch 3/30\n",
            "313/313 [==============================] - 65s 208ms/step - loss: 0.1564 - accuracy: 0.9404 - val_loss: 0.0520 - val_accuracy: 0.9737 - lr: 0.0010\n",
            "Epoch 4/30\n",
            "313/313 [==============================] - 65s 207ms/step - loss: 0.1309 - accuracy: 0.9517 - val_loss: 0.0537 - val_accuracy: 0.9787 - lr: 0.0010\n",
            "Epoch 5/30\n",
            "313/313 [==============================] - ETA: 0s - loss: 0.1170 - accuracy: 0.9581\n",
            "Epoch 00005: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.\n",
            "313/313 [==============================] - 65s 207ms/step - loss: 0.1170 - accuracy: 0.9581 - val_loss: 0.2111 - val_accuracy: 0.9025 - lr: 0.0010\n",
            "Epoch 6/30\n",
            "313/313 [==============================] - 65s 208ms/step - loss: 0.0855 - accuracy: 0.9697 - val_loss: 0.0403 - val_accuracy: 0.9862 - lr: 1.0000e-04\n",
            "Epoch 7/30\n",
            "313/313 [==============================] - 65s 208ms/step - loss: 0.0844 - accuracy: 0.9706 - val_loss: 0.0342 - val_accuracy: 0.9887 - lr: 1.0000e-04\n",
            "Epoch 8/30\n",
            "313/313 [==============================] - 65s 207ms/step - loss: 0.0806 - accuracy: 0.9709 - val_loss: 0.0271 - val_accuracy: 0.9887 - lr: 1.0000e-04\n",
            "Epoch 9/30\n",
            "313/313 [==============================] - 66s 210ms/step - loss: 0.0789 - accuracy: 0.9720 - val_loss: 0.0299 - val_accuracy: 0.9925 - lr: 1.0000e-04\n",
            "Epoch 10/30\n",
            "313/313 [==============================] - ETA: 0s - loss: 0.0735 - accuracy: 0.9731\n",
            "Epoch 00010: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.\n",
            "313/313 [==============================] - 65s 207ms/step - loss: 0.0735 - accuracy: 0.9731 - val_loss: 0.0380 - val_accuracy: 0.9887 - lr: 1.0000e-04\n",
            "Epoch 11/30\n",
            "313/313 [==============================] - 65s 207ms/step - loss: 0.0705 - accuracy: 0.9750 - val_loss: 0.0281 - val_accuracy: 0.9912 - lr: 1.0000e-05\n",
            "Epoch 12/30\n",
            "313/313 [==============================] - ETA: 0s - loss: 0.0700 - accuracy: 0.9754\n",
            "Epoch 00012: ReduceLROnPlateau reducing learning rate to 1.0000000656873453e-06.\n",
            "313/313 [==============================] - 65s 207ms/step - loss: 0.0700 - accuracy: 0.9754 - val_loss: 0.0279 - val_accuracy: 0.9900 - lr: 1.0000e-05\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "O7bWbIJwqRkM",
        "outputId": "948a7a95-7984-4612-a472-60cdf25c39ac"
      },
      "source": [
        "# Visulizing the loss functions\n",
        "import matplotlib.pyplot as plt\n",
        "plt.plot(history.history['loss'],label=\"Training loss\")\n",
        "plt.plot(history.history['val_loss'],label=\"Validation loss\")\n",
        "plt.xlabel('epochs')\n",
        "plt.ylabel('loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "ZKNCV95Jq2c0",
        "outputId": "3f7b1518-7aeb-4964-973f-b765b19db4da"
      },
      "source": [
        "# Visualizing the accuracy of model\n",
        "plt.plot(history.history['accuracy'],label=\"Training accuracy\")\n",
        "plt.plot(history.history['val_accuracy'],label=\"Validation accuracy\")\n",
        "plt.xlabel('epochs')\n",
        "plt.ylabel('loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xKJYNTS5AHTr"
      },
      "source": [
        "# Saving the model\n",
        "model.save(\"model.h5\",model)"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}