drowsiness_detection.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "91579fe7",
      "metadata": {
        "papermill": {
          "duration": 0.008422,
          "end_time": "2023-03-24T09:18:32.859777",
          "exception": false,
          "start_time": "2023-03-24T09:18:32.851355",
          "status": "completed"
        },
        "tags": [],
        "id": "91579fe7"
      },
      "source": [
        "# **Importing the libraries**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "id": "9a71493c",
      "metadata": {
        "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
        "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
        "execution": {
          "iopub.execute_input": "2023-03-24T09:18:32.875452Z",
          "iopub.status.busy": "2023-03-24T09:18:32.875019Z",
          "iopub.status.idle": "2023-03-24T09:18:41.302798Z",
          "shell.execute_reply": "2023-03-24T09:18:41.301240Z"
        },
        "papermill": {
          "duration": 8.438967,
          "end_time": "2023-03-24T09:18:41.305882",
          "exception": false,
          "start_time": "2023-03-24T09:18:32.866915",
          "status": "completed"
        },
        "tags": [],
        "id": "9a71493c"
      },
      "outputs": [],
      "source": [
        "import numpy as np \n",
        "import pandas as pd \n",
        "import os\n",
        "import tensorflow as tf\n",
        "\n",
        "from glob import glob\n",
        "from tqdm import tqdm\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "from PIL import Image"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "769a64ce",
      "metadata": {
        "papermill": {
          "duration": 0.016082,
          "end_time": "2023-03-24T09:18:41.328852",
          "exception": false,
          "start_time": "2023-03-24T09:18:41.312770",
          "status": "completed"
        },
        "tags": [],
        "id": "769a64ce"
      },
      "source": [
        "#  **Loading the Images**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "id": "03fbe3d1",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:18:41.346130Z",
          "iopub.status.busy": "2023-03-24T09:18:41.344654Z",
          "iopub.status.idle": "2023-03-24T09:19:11.971418Z",
          "shell.execute_reply": "2023-03-24T09:19:11.970095Z"
        },
        "papermill": {
          "duration": 30.639378,
          "end_time": "2023-03-24T09:19:11.975552",
          "exception": false,
          "start_time": "2023-03-24T09:18:41.336174",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "03fbe3d1",
        "outputId": "25ca412d-8b96-492d-cd31-ba99acdeb9a5"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 2000/2000 [00:05<00:00, 335.44it/s]\n",
            "100%|██████████| 2000/2000 [00:04<00:00, 420.76it/s]\n"
          ]
        }
      ],
      "source": [
        "X = []\n",
        "Y = []\n",
        "\n",
        "for i in tqdm(glob('/content/drive/MyDrive/Shared/Drowsiness Detection/train/Open_Eyes/*')):\n",
        "    temp = np.array(Image.open(i).resize((64,64)))\n",
        "    X.append(temp)\n",
        "    Y.append(1)\n",
        "    \n",
        "for i in tqdm(glob('/content/drive/MyDrive/Shared/Drowsiness Detection/train/Closed_Eyes/*')):\n",
        "    temp = np.array(Image.open(i).resize((64,64)))\n",
        "    X.append(temp)\n",
        "    Y.append(0)    "
      ]
    },
    {
      "cell_type": "markdown",
      "id": "4bf35493",
      "metadata": {
        "papermill": {
          "duration": 0.024194,
          "end_time": "2023-03-24T09:19:12.025211",
          "exception": false,
          "start_time": "2023-03-24T09:19:12.001017",
          "status": "completed"
        },
        "tags": [],
        "id": "4bf35493"
      },
      "source": [
        "# **Normalizing the images (with any algorithm)**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "id": "e292f3a5",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:19:12.076077Z",
          "iopub.status.busy": "2023-03-24T09:19:12.074954Z",
          "iopub.status.idle": "2023-03-24T09:19:12.080200Z",
          "shell.execute_reply": "2023-03-24T09:19:12.079262Z"
        },
        "papermill": {
          "duration": 0.033625,
          "end_time": "2023-03-24T09:19:12.082626",
          "exception": false,
          "start_time": "2023-03-24T09:19:12.049001",
          "status": "completed"
        },
        "tags": [],
        "id": "e292f3a5"
      },
      "outputs": [],
      "source": [
        "#  X = np.array(X)\n",
        "# X = X/255.0\n",
        "# Y = np.array(Y)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "16a86f4e",
      "metadata": {
        "papermill": {
          "duration": 0.024529,
          "end_time": "2023-03-24T09:19:12.130533",
          "exception": false,
          "start_time": "2023-03-24T09:19:12.106004",
          "status": "completed"
        },
        "tags": [],
        "id": "16a86f4e"
      },
      "source": [
        "# **Normalizing the images using Min-Max Scaling algorithm**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "id": "e2d0e082",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:19:12.181629Z",
          "iopub.status.busy": "2023-03-24T09:19:12.180509Z",
          "iopub.status.idle": "2023-03-24T09:19:12.396796Z",
          "shell.execute_reply": "2023-03-24T09:19:12.395426Z"
        },
        "papermill": {
          "duration": 0.245338,
          "end_time": "2023-03-24T09:19:12.399694",
          "exception": false,
          "start_time": "2023-03-24T09:19:12.154356",
          "status": "completed"
        },
        "tags": [],
        "id": "e2d0e082"
      },
      "outputs": [],
      "source": [
        "X = (np.array(X) - np.min(X)) / (np.max(X) - np.min(X))\n",
        "X = X/255.0\n",
        "Y = (np.array(Y) - np.min(Y)) / (np.max(Y) - np.min(Y))"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "0ceb03dd",
      "metadata": {
        "papermill": {
          "duration": 0.0234,
          "end_time": "2023-03-24T09:19:12.447162",
          "exception": false,
          "start_time": "2023-03-24T09:19:12.423762",
          "status": "completed"
        },
        "tags": [],
        "id": "0ceb03dd"
      },
      "source": [
        "# **Normalizing the images using Z-Score algorithm**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "id": "902e6f89",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:19:12.497631Z",
          "iopub.status.busy": "2023-03-24T09:19:12.497194Z",
          "iopub.status.idle": "2023-03-24T09:19:12.502336Z",
          "shell.execute_reply": "2023-03-24T09:19:12.501061Z"
        },
        "papermill": {
          "duration": 0.033572,
          "end_time": "2023-03-24T09:19:12.505202",
          "exception": false,
          "start_time": "2023-03-24T09:19:12.471630",
          "status": "completed"
        },
        "tags": [],
        "id": "902e6f89"
      },
      "outputs": [],
      "source": [
        "# 'X = (np.array(X) - np.mean(X)) / np.std(X)\n",
        "# X = X/255.0\n",
        "# Y = (np.array(Y) - np.mean(Y)) / np.std(Y)'"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "06bc370d",
      "metadata": {
        "papermill": {
          "duration": 0.023503,
          "end_time": "2023-03-24T09:19:12.551867",
          "exception": false,
          "start_time": "2023-03-24T09:19:12.528364",
          "status": "completed"
        },
        "tags": [],
        "id": "06bc370d"
      },
      "source": [
        "# Expanding the dimensions of array (numpy)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "id": "493f1dd5",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:19:12.616685Z",
          "iopub.status.busy": "2023-03-24T09:19:12.615677Z",
          "iopub.status.idle": "2023-03-24T09:19:12.621678Z",
          "shell.execute_reply": "2023-03-24T09:19:12.620480Z"
        },
        "papermill": {
          "duration": 0.044992,
          "end_time": "2023-03-24T09:19:12.624904",
          "exception": false,
          "start_time": "2023-03-24T09:19:12.579912",
          "status": "completed"
        },
        "tags": [],
        "id": "493f1dd5"
      },
      "outputs": [],
      "source": [
        "X = np.expand_dims(X,-1)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "6406ed0e",
      "metadata": {
        "papermill": {
          "duration": 0.035265,
          "end_time": "2023-03-24T09:19:12.697198",
          "exception": false,
          "start_time": "2023-03-24T09:19:12.661933",
          "status": "completed"
        },
        "tags": [],
        "id": "6406ed0e"
      },
      "source": [
        "# Spliting the dataset into training and testing data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "id": "f9a63eda",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:19:12.752657Z",
          "iopub.status.busy": "2023-03-24T09:19:12.752218Z",
          "iopub.status.idle": "2023-03-24T09:19:13.460245Z",
          "shell.execute_reply": "2023-03-24T09:19:13.459208Z"
        },
        "papermill": {
          "duration": 0.737718,
          "end_time": "2023-03-24T09:19:13.463001",
          "exception": false,
          "start_time": "2023-03-24T09:19:12.725283",
          "status": "completed"
        },
        "tags": [],
        "id": "f9a63eda"
      },
      "outputs": [],
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "x_train, x_test, y_train, y_test = train_test_split(X, Y, test_size=0.33, random_state=42)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "0682f86c",
      "metadata": {
        "papermill": {
          "duration": 0.026472,
          "end_time": "2023-03-24T09:19:13.514034",
          "exception": false,
          "start_time": "2023-03-24T09:19:13.487562",
          "status": "completed"
        },
        "tags": [],
        "id": "0682f86c"
      },
      "source": [
        "# CNN Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "id": "f5b41ec3",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:19:13.563318Z",
          "iopub.status.busy": "2023-03-24T09:19:13.562511Z",
          "iopub.status.idle": "2023-03-24T09:19:14.136049Z",
          "shell.execute_reply": "2023-03-24T09:19:14.135079Z"
        },
        "papermill": {
          "duration": 0.601002,
          "end_time": "2023-03-24T09:19:14.138585",
          "exception": false,
          "start_time": "2023-03-24T09:19:13.537583",
          "status": "completed"
        },
        "tags": [],
        "id": "f5b41ec3"
      },
      "outputs": [],
      "source": [
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense, Input, Conv2D, BatchNormalization, MaxPooling2D,Dropout, Flatten\n",
        "\n",
        "model = tf.keras.models.Sequential([\n",
        "      Input(shape=(64, 64, 1)),\n",
        "\n",
        "      Conv2D(filters = 32, kernel_size = 5, strides = 1, activation = 'relu'),\n",
        "      Conv2D(filters = 32, kernel_size = 5, strides = 1, activation = 'relu', use_bias=False),\n",
        "      BatchNormalization(),\n",
        "      MaxPooling2D(strides = 2),\n",
        "      Dropout(0.3),\n",
        "\n",
        "      Conv2D(filters = 64, kernel_size = 3, strides = 1, activation = 'relu'),\n",
        "      Conv2D(filters = 64, kernel_size = 3, strides = 1, activation = 'relu', use_bias=False),\n",
        "      BatchNormalization(),\n",
        "      MaxPooling2D(strides = 2),\n",
        "      Dropout(0.3),\n",
        "\n",
        "      Flatten(),\n",
        "      Dense(units  = 256, activation = 'relu', use_bias=False),\n",
        "      BatchNormalization(),\n",
        "\n",
        "      Dense(units = 128, use_bias=False, activation = 'relu'),\n",
        "\n",
        "      Dense(units = 84, use_bias=False, activation = 'relu'),\n",
        "      BatchNormalization(),\n",
        "      Dropout(0.3),\n",
        "\n",
        "      Dense(units = 1, activation = 'sigmoid')\n",
        "  ])"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "c80edfc4",
      "metadata": {
        "papermill": {
          "duration": 0.024465,
          "end_time": "2023-03-24T09:19:14.186980",
          "exception": false,
          "start_time": "2023-03-24T09:19:14.162515",
          "status": "completed"
        },
        "tags": [],
        "id": "c80edfc4"
      },
      "source": [
        "# Compiling the Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "id": "42643b1a",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:19:14.237773Z",
          "iopub.status.busy": "2023-03-24T09:19:14.237309Z",
          "iopub.status.idle": "2023-03-24T09:19:14.258237Z",
          "shell.execute_reply": "2023-03-24T09:19:14.256867Z"
        },
        "papermill": {
          "duration": 0.048673,
          "end_time": "2023-03-24T09:19:14.261103",
          "exception": false,
          "start_time": "2023-03-24T09:19:14.212430",
          "status": "completed"
        },
        "tags": [],
        "id": "42643b1a"
      },
      "outputs": [],
      "source": [
        "model.compile(loss='binary_crossentropy',optimizer='adam', metrics=['accuracy'])"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "a79c38a6",
      "metadata": {
        "papermill": {
          "duration": 0.024359,
          "end_time": "2023-03-24T09:19:14.309981",
          "exception": false,
          "start_time": "2023-03-24T09:19:14.285622",
          "status": "completed"
        },
        "tags": [],
        "id": "a79c38a6"
      },
      "source": [
        "# Training of the Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 37,
      "id": "2a0c5c55",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:19:14.360265Z",
          "iopub.status.busy": "2023-03-24T09:19:14.359835Z",
          "iopub.status.idle": "2023-03-24T09:37:38.868746Z",
          "shell.execute_reply": "2023-03-24T09:37:38.867393Z"
        },
        "papermill": {
          "duration": 1104.69597,
          "end_time": "2023-03-24T09:37:39.030153",
          "exception": false,
          "start_time": "2023-03-24T09:19:14.334183",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2a0c5c55",
        "outputId": "e5450070-fd9c-41ee-c49d-81aa3385788f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "67/67 [==============================] - ETA: 0s - loss: 0.1288 - accuracy: 0.9534\n",
            "Epoch 1: val_loss improved from inf to 0.68563, saving model to bestModel.h5\n",
            "67/67 [==============================] - 79s 1s/step - loss: 0.1288 - accuracy: 0.9534 - val_loss: 0.6856 - val_accuracy: 0.9590\n",
            "Epoch 2/10\n",
            "67/67 [==============================] - ETA: 0s - loss: 0.0548 - accuracy: 0.9832\n",
            "Epoch 2: val_loss did not improve from 0.68563\n",
            "67/67 [==============================] - 76s 1s/step - loss: 0.0548 - accuracy: 0.9832 - val_loss: 1.0363 - val_accuracy: 0.4813\n",
            "Epoch 3/10\n",
            "67/67 [==============================] - ETA: 0s - loss: 0.0497 - accuracy: 0.9841\n",
            "Epoch 3: val_loss did not improve from 0.68563\n",
            "67/67 [==============================] - 72s 1s/step - loss: 0.0497 - accuracy: 0.9841 - val_loss: 0.7480 - val_accuracy: 0.4813\n",
            "Epoch 4/10\n",
            "67/67 [==============================] - ETA: 0s - loss: 0.0377 - accuracy: 0.9888\n",
            "Epoch 4: val_loss did not improve from 0.68563\n",
            "67/67 [==============================] - 75s 1s/step - loss: 0.0377 - accuracy: 0.9888 - val_loss: 0.7727 - val_accuracy: 0.4813\n",
            "Epoch 5/10\n",
            "67/67 [==============================] - ETA: 0s - loss: 0.0130 - accuracy: 0.9967\n",
            "Epoch 5: val_loss improved from 0.68563 to 0.59139, saving model to bestModel.h5\n",
            "67/67 [==============================] - 76s 1s/step - loss: 0.0130 - accuracy: 0.9967 - val_loss: 0.5914 - val_accuracy: 0.6679\n",
            "Epoch 6/10\n",
            "67/67 [==============================] - ETA: 0s - loss: 0.0256 - accuracy: 0.9921\n",
            "Epoch 6: val_loss improved from 0.59139 to 0.27345, saving model to bestModel.h5\n",
            "67/67 [==============================] - 75s 1s/step - loss: 0.0256 - accuracy: 0.9921 - val_loss: 0.2734 - val_accuracy: 0.9198\n",
            "Epoch 7/10\n",
            "67/67 [==============================] - ETA: 0s - loss: 0.0170 - accuracy: 0.9949\n",
            "Epoch 7: val_loss did not improve from 0.27345\n",
            "67/67 [==============================] - 75s 1s/step - loss: 0.0170 - accuracy: 0.9949 - val_loss: 0.7446 - val_accuracy: 0.6978\n",
            "Epoch 8/10\n",
            "67/67 [==============================] - ETA: 0s - loss: 0.0213 - accuracy: 0.9939\n",
            "Epoch 8: val_loss improved from 0.27345 to 0.10015, saving model to bestModel.h5\n",
            "67/67 [==============================] - 74s 1s/step - loss: 0.0213 - accuracy: 0.9939 - val_loss: 0.1002 - val_accuracy: 0.9832\n",
            "Epoch 9/10\n",
            "67/67 [==============================] - ETA: 0s - loss: 0.0157 - accuracy: 0.9963\n",
            "Epoch 9: val_loss did not improve from 0.10015\n",
            "67/67 [==============================] - 75s 1s/step - loss: 0.0157 - accuracy: 0.9963 - val_loss: 0.8230 - val_accuracy: 0.6866\n",
            "Epoch 10/10\n",
            "67/67 [==============================] - ETA: 0s - loss: 0.0148 - accuracy: 0.9963\n",
            "Epoch 10: val_loss did not improve from 0.10015\n",
            "67/67 [==============================] - 75s 1s/step - loss: 0.0148 - accuracy: 0.9963 - val_loss: 0.6151 - val_accuracy: 0.7537\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7faa105e9870>"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ],
      "source": [
        "callback = tf.keras.callbacks.ModelCheckpoint(\n",
        "    filepath='bestModel.h5',\n",
        "    save_weights_only=False,\n",
        "    monitor='val_loss',\n",
        "    mode='min',\n",
        "    save_best_only=True,\n",
        "    verbose =1)\n",
        "model.fit(x_train, y_train, validation_split=0.2, epochs=10, batch_size=32, callbacks=callback)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "30b616da",
      "metadata": {
        "papermill": {
          "duration": 0.158583,
          "end_time": "2023-03-24T09:37:39.348845",
          "exception": false,
          "start_time": "2023-03-24T09:37:39.190262",
          "status": "completed"
        },
        "tags": [],
        "id": "30b616da"
      },
      "source": [
        "# Model Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 38,
      "id": "77f5a46e",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:37:39.670732Z",
          "iopub.status.busy": "2023-03-24T09:37:39.670265Z",
          "iopub.status.idle": "2023-03-24T09:37:43.725321Z",
          "shell.execute_reply": "2023-03-24T09:37:43.724073Z"
        },
        "papermill": {
          "duration": 4.219052,
          "end_time": "2023-03-24T09:37:43.728118",
          "exception": false,
          "start_time": "2023-03-24T09:37:39.509066",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "77f5a46e",
        "outputId": "7428acbc-563c-438d-bd1f-32bc743823e7"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "42/42 [==============================] - 10s 240ms/step - loss: 0.5230 - accuracy: 0.7902\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.5230398178100586, 0.7901515364646912]"
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ],
      "source": [
        "model.evaluate(x_test, y_test)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "8cd5b64e",
      "metadata": {
        "papermill": {
          "duration": 0.163336,
          "end_time": "2023-03-24T09:37:44.052637",
          "exception": false,
          "start_time": "2023-03-24T09:37:43.889301",
          "status": "completed"
        },
        "tags": [],
        "id": "8cd5b64e"
      },
      "source": [
        "# Loading the Model Performance"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 39,
      "id": "f1bd6724",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:37:44.438123Z",
          "iopub.status.busy": "2023-03-24T09:37:44.437465Z",
          "iopub.status.idle": "2023-03-24T09:37:50.165904Z",
          "shell.execute_reply": "2023-03-24T09:37:50.164709Z"
        },
        "papermill": {
          "duration": 5.953312,
          "end_time": "2023-03-24T09:37:50.168739",
          "exception": false,
          "start_time": "2023-03-24T09:37:44.215427",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "f1bd6724",
        "outputId": "39208a09-55e6-46e8-e2bc-1d9d12e3150b"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "42/42 [==============================] - 10s 231ms/step - loss: 0.1033 - accuracy: 0.9803\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.10332013666629791, 0.9803030490875244]"
            ]
          },
          "metadata": {},
          "execution_count": 39
        }
      ],
      "source": [
        "from keras.models import load_model\n",
        "best_model = load_model('/content/bestModel.h5')\n",
        "best_model.evaluate(x_test, y_test)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "58358a31",
      "metadata": {
        "papermill": {
          "duration": 0.162072,
          "end_time": "2023-03-24T09:37:50.496065",
          "exception": false,
          "start_time": "2023-03-24T09:37:50.333993",
          "status": "completed"
        },
        "tags": [],
        "id": "58358a31"
      },
      "source": [
        "# Changing the dimensions and Plotting the trained images"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "id": "b6c7a8de",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:37:50.825959Z",
          "iopub.status.busy": "2023-03-24T09:37:50.825552Z",
          "iopub.status.idle": "2023-03-24T09:37:52.440932Z",
          "shell.execute_reply": "2023-03-24T09:37:52.438943Z"
        },
        "papermill": {
          "duration": 1.78554,
          "end_time": "2023-03-24T09:37:52.443970",
          "exception": false,
          "start_time": "2023-03-24T09:37:50.658430",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "b6c7a8de",
        "outputId": "bb1330f8-85a4-4395-dcd5-5ef1b0ee8cf6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 234ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Open\n",
            "1/1 [==============================] - 0s 47ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Closed\n",
            "1/1 [==============================] - 0s 30ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Open\n",
            "1/1 [==============================] - 0s 29ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Closed\n",
            "1/1 [==============================] - 0s 31ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Closed\n"
          ]
        }
      ],
      "source": [
        "for i in x_test[0:5]:\n",
        "    result = best_model.predict(np.expand_dims(i,0))\n",
        "    plt.imshow(i)\n",
        "    plt.show()\n",
        "    \n",
        "    if result > 0.5:\n",
        "        print('Open')\n",
        "    else:\n",
        "        print(\"Closed\")\n",
        "    "
      ]
    },
    {
      "cell_type": "markdown",
      "id": "be57beab",
      "metadata": {
        "papermill": {
          "duration": 0.166873,
          "end_time": "2023-03-24T09:37:52.776532",
          "exception": false,
          "start_time": "2023-03-24T09:37:52.609659",
          "status": "completed"
        },
        "tags": [],
        "id": "be57beab"
      },
      "source": [
        "# Prediction and Confusion Matrix"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 41,
      "id": "b6d40124",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:37:53.111781Z",
          "iopub.status.busy": "2023-03-24T09:37:53.110500Z",
          "iopub.status.idle": "2023-03-24T09:37:58.204937Z",
          "shell.execute_reply": "2023-03-24T09:37:58.203739Z"
        },
        "papermill": {
          "duration": 5.263951,
          "end_time": "2023-03-24T09:37:58.207597",
          "exception": false,
          "start_time": "2023-03-24T09:37:52.943646",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 543
        },
        "id": "b6d40124",
        "outputId": "0c3659e4-6dc1-408c-c05a-e1a8b157fdcc"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "42/42 [==============================] - 8s 182ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 25.722222222222214, 'Actual')"
            ]
          },
          "metadata": {},
          "execution_count": 41
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1500x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "import seaborn as sns\n",
        "\n",
        "plt.figure(figsize=(15, 5))\n",
        "\n",
        "preds = best_model.predict(x_test)\n",
        "preds = (preds >= 0.5).astype(np.int32)\n",
        "cm = confusion_matrix(y_test, preds)\n",
        "df_cm = pd.DataFrame(cm, index=['closed', 'Open'], columns=['Closed', 'Open'])\n",
        "plt.subplot(121)\n",
        "plt.title(\"Confusion matrix\\n\")\n",
        "sns.heatmap(df_cm, annot=True, fmt=\"d\", cmap=\"YlGnBu\")\n",
        "plt.ylabel(\"Predicted\")\n",
        "plt.xlabel(\"Actual\")"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "e7cd3f3e",
      "metadata": {
        "papermill": {
          "duration": 0.168578,
          "end_time": "2023-03-24T09:37:58.545174",
          "exception": false,
          "start_time": "2023-03-24T09:37:58.376596",
          "status": "completed"
        },
        "tags": [],
        "id": "e7cd3f3e"
      },
      "source": [
        "# Accuracy\n",
        "\n",
        "Evaluating the Accuracy"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 42,
      "id": "40362445",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2023-03-24T09:37:58.887103Z",
          "iopub.status.busy": "2023-03-24T09:37:58.886441Z",
          "iopub.status.idle": "2023-03-24T09:38:03.029123Z",
          "shell.execute_reply": "2023-03-24T09:38:03.027783Z"
        },
        "papermill": {
          "duration": 4.316324,
          "end_time": "2023-03-24T09:38:03.031758",
          "exception": false,
          "start_time": "2023-03-24T09:37:58.715434",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "40362445",
        "outputId": "6e3a142f-1c1d-4749-aa72-8d533c0a9b6e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "42/42 [==============================] - 10s 240ms/step - loss: 0.1033 - accuracy: 0.9803\n",
            "Accuracy: 0.9803030490875244\n"
          ]
        }
      ],
      "source": [
        "accuracy = best_model.evaluate(x_test, y_test)[1]\n",
        "print(\"Accuracy:\", accuracy)"
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.7.12"
    },
    "papermill": {
      "default_parameters": {},
      "duration": 1183.507204,
      "end_time": "2023-03-24T09:38:05.948610",
      "environment_variables": {},
      "exception": null,
      "input_path": "__notebook__.ipynb",
      "output_path": "__notebook__.ipynb",
      "parameters": {},
      "start_time": "2023-03-24T09:18:22.441406",
      "version": "2.4.0"
    },
    "colab": {
      "provenance": []
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}