Mood_Detection_ML_Project.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "gpuClass": "standard"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/jabhij/MusicTherapyTreatment_EmotionDetection/blob/main/Mood_Detection_ML_Project.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "VpdU9gG-qH8b"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import os\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.metrics import confusion_matrix\n",
        "\n",
        "from keras import models\n",
        "from keras.utils import to_categorical, np_utils\n",
        "\n",
        "from tensorflow import convert_to_tensor\n",
        "\n",
        "from tensorflow.keras.layers import Conv2D, Flatten, Dense, MaxPool2D, BatchNormalization, ZeroPadding2D, Dropout\n",
        "from tensorflow.keras import models\n",
        "from tensorflow.keras.callbacks import Callback, EarlyStopping, ReduceLROnPlateau\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from tensorflow.keras.utils import plot_model\n",
        "import tensorflow_datasets as tfd"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "id": "Ig6rzFGDtQLN",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "01a99a49-2803-4ae6-8acb-e44b781bbf61"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "path = '/content/drive/MyDrive/Mood-Detection-Project/challenges-in-representation-learning-facial-expression-recognition-challenge/'"
      ],
      "metadata": {
        "id": "Ygw4gSW9zAs_"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data = pd.read_csv(path + 'icml_face_data.csv')"
      ],
      "metadata": {
        "id": "D8Kg0NQU4jqF"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data.head()"
      ],
      "metadata": {
        "id": "sKqyUP924xz-",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "b496c0eb-f2fe-4fe4-ca31-17ad216bbf71"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   emotion     Usage                                             pixels\n",
              "0        0  Training  70 80 82 72 58 58 60 63 54 58 60 48 89 115 121...\n",
              "1        0  Training  151 150 147 155 148 133 111 140 170 174 182 15...\n",
              "2        2  Training  231 212 156 164 174 138 161 173 182 200 106 38...\n",
              "3        4  Training  24 32 36 30 32 23 19 20 30 41 21 22 32 34 21 1...\n",
              "4        6  Training  4 0 0 0 0 0 0 0 0 0 0 0 3 15 23 28 48 50 58 84..."
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f6dcd6e1-76fb-474f-8ccb-d043505231cb\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>emotion</th>\n",
              "      <th>Usage</th>\n",
              "      <th>pixels</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>Training</td>\n",
              "      <td>70 80 82 72 58 58 60 63 54 58 60 48 89 115 121...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>Training</td>\n",
              "      <td>151 150 147 155 148 133 111 140 170 174 182 15...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>Training</td>\n",
              "      <td>231 212 156 164 174 138 161 173 182 200 106 38...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>Training</td>\n",
              "      <td>24 32 36 30 32 23 19 20 30 41 21 22 32 34 21 1...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>6</td>\n",
              "      <td>Training</td>\n",
              "      <td>4 0 0 0 0 0 0 0 0 0 0 0 3 15 23 28 48 50 58 84...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f6dcd6e1-76fb-474f-8ccb-d043505231cb')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-f6dcd6e1-76fb-474f-8ccb-d043505231cb button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-f6dcd6e1-76fb-474f-8ccb-d043505231cb');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# filter rows with emotion == 1\n",
        "emotion_1 = data[data['emotion'] == 1]\n",
        "\n",
        "# repeat rows 5 times\n",
        "repeated_rows = pd.concat([emotion_1] * 6, ignore_index=True)\n",
        "\n",
        "#concating augmented data and making a new dataframe \n",
        "data2 = pd.concat([data, repeated_rows], ignore_index=True)\n"
      ],
      "metadata": {
        "id": "WSQXd1lANxnd"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data2.info()"
      ],
      "metadata": {
        "id": "CM4IofbL2InM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ea1cdd23-e833-4a89-8bd1-eb33ce4e3dbd"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 39169 entries, 0 to 39168\n",
            "Data columns (total 3 columns):\n",
            " #   Column   Non-Null Count  Dtype \n",
            "---  ------   --------------  ----- \n",
            " 0   emotion  39169 non-null  int64 \n",
            " 1    Usage   39169 non-null  object\n",
            " 2    pixels  39169 non-null  object\n",
            "dtypes: int64(1), object(2)\n",
            "memory usage: 918.1+ KB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data2[data2['emotion'] == 1].count()"
      ],
      "metadata": {
        "id": "Wf8UXnVBDgep",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f412bddf-306b-4cd5-9aff-d69a7fd1d7ae"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "emotion    3829\n",
              " Usage     3829\n",
              " pixels    3829\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data2.info()"
      ],
      "metadata": {
        "id": "iwTaCfAp5U_-",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0d9805db-3b2b-4448-f8bf-cde7ece01baf"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 39169 entries, 0 to 39168\n",
            "Data columns (total 3 columns):\n",
            " #   Column   Non-Null Count  Dtype \n",
            "---  ------   --------------  ----- \n",
            " 0   emotion  39169 non-null  int64 \n",
            " 1    Usage   39169 non-null  object\n",
            " 2    pixels  39169 non-null  object\n",
            "dtypes: int64(1), object(2)\n",
            "memory usage: 918.1+ KB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def prepare_data(data):\n",
        "    \"\"\" Prepare data for modeling \n",
        "        input: data frame with labels und pixel data\n",
        "        output: image and label array \"\"\"\n",
        "    \n",
        "    image_array = np.zeros(shape=(len(data), 48, 48))\n",
        "    image_label = np.array(list(map(int, data['emotion'])))\n",
        "    \n",
        "    for i, row in enumerate(data.index):\n",
        "        image = np.fromstring(data.loc[row, ' pixels'], dtype=int, sep=' ')\n",
        "        image = np.reshape(image, (48, 48))\n",
        "        image_array[i] = image\n",
        "        \n",
        "    return image_array, image_label\n",
        "\n",
        "\n",
        "def plot_examples(label=0):\n",
        "    fig, axs = plt.subplots(1, 5, figsize=(25, 12))\n",
        "    fig.subplots_adjust(hspace = .2, wspace=.2)\n",
        "    axs = axs.ravel()\n",
        "    for i in range(5):\n",
        "        idx = data[data['emotion']==label].index[i]\n",
        "        axs[i].imshow(train_images[idx][:,:,0], cmap='gray')\n",
        "        axs[i].set_title(emotions[train_labels[idx].argmax()])\n",
        "        axs[i].set_xticklabels([])\n",
        "        axs[i].set_yticklabels([])\n",
        "        \n",
        "def plot_all_emotions():\n",
        "    fig, axs = plt.subplots(1, 7, figsize=(30, 12))\n",
        "    fig.subplots_adjust(hspace = .2, wspace=.2)\n",
        "    axs = axs.ravel()\n",
        "    for i in range(7):\n",
        "        idx = data[data['emotion']==i].index[i]\n",
        "        axs[i].imshow(train_images[idx][:,:,0], cmap='gray')\n",
        "        axs[i].set_title(emotions[train_labels[idx].argmax()])\n",
        "        axs[i].set_xticklabels([])\n",
        "        axs[i].set_yticklabels([])\n",
        "        \n",
        "def plot_image_and_emotion(test_image_array, test_image_label, pred_test_labels, image_number):\n",
        "    \"\"\" Function to plot the image and compare the prediction results with the label \"\"\"\n",
        "    \n",
        "    fig, axs = plt.subplots(1, 2, figsize=(12, 6), sharey=False)\n",
        "    \n",
        "    bar_label = emotions.values()\n",
        "    \n",
        "    axs[0].imshow(test_image_array[image_number], 'gray')\n",
        "    axs[0].set_title(emotions[test_image_label[image_number]])\n",
        "    \n",
        "    axs[1].bar(bar_label, pred_test_labels[image_number], color='orange', alpha=0.7)\n",
        "    axs[1].grid()\n",
        "    \n",
        "    plt.show()\n",
        "    \n",
        "def plot_compare_distributions(array1, array2, title1='', title2=''):\n",
        "    df_array1 = pd.DataFrame()\n",
        "    df_array2 = pd.DataFrame()\n",
        "    df_array1['emotion'] = array1.argmax(axis=1)\n",
        "    df_array2['emotion'] = array2.argmax(axis=1)\n",
        "    \n",
        "    fig, axs = plt.subplots(1, 2, figsize=(12, 6), sharey=False)\n",
        "    x = emotions.values()\n",
        "    \n",
        "    y = df_array1['emotion'].value_counts()\n",
        "    keys_missed = list(set(emotions.keys()).difference(set(y.keys())))\n",
        "    for key_missed in keys_missed:\n",
        "        y[key_missed] = 0\n",
        "    axs[0].bar(x, y.sort_index(), color='orange')\n",
        "    axs[0].set_title(title1)\n",
        "    axs[0].grid()\n",
        "    \n",
        "    y = df_array2['emotion'].value_counts()\n",
        "    keys_missed = list(set(emotions.keys()).difference(set(y.keys())))\n",
        "    for key_missed in keys_missed:\n",
        "        y[key_missed] = 0\n",
        "    axs[1].bar(x, y.sort_index())\n",
        "    axs[1].set_title(title2)\n",
        "    axs[1].grid()\n",
        "    \n",
        "    plt.show()"
      ],
      "metadata": {
        "id": "fI985AaM6ICI"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "emotions = {0: 'Angry', 1: 'Disgust', 2: 'Fear', 3: 'Happy', 4: 'Sad', 5: 'Surprise', 6: 'Neutral'}"
      ],
      "metadata": {
        "id": "7-QK7jGG6lVV"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_image_array, train_image_label = prepare_data(data2[data2[' Usage']=='Training'])\n",
        "val_image_array, val_image_label = prepare_data(data2[data2[' Usage']=='PrivateTest'])\n",
        "test_image_array, test_image_label = prepare_data(data2[data2[' Usage']=='PublicTest'])"
      ],
      "metadata": {
        "id": "rpEUpjya6lcB"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_images = train_image_array.reshape((train_image_array.shape[0], 48, 48, 1))\n",
        "train_images = train_images.astype('float32')/255\n",
        "val_images = val_image_array.reshape((val_image_array.shape[0], 48,48,1))\n",
        "val_images = val_images.astype('float32')/255\n",
        "test_images = test_image_array.reshape((test_image_array.shape[0], 48, 48, 1))\n",
        "test_images = test_images.astype('float32')/255"
      ],
      "metadata": {
        "id": "3nXDdFSe6lg1"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_labels = to_categorical(train_image_label)\n",
        "val_labels = to_categorical(val_image_label)\n",
        "test_labels = to_categorical(test_image_label)"
      ],
      "metadata": {
        "id": "24l2QdSH6llV"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_all_emotions()"
      ],
      "metadata": {
        "id": "ws6oS84s7FNZ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 168
        },
        "outputId": "37c94f69-b29a-4c57-cf94-376de0a3e332"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 3000x1200 with 7 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_examples(label=1)"
      ],
      "metadata": {
        "id": "l23cvsGQ7Rn9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 226
        },
        "outputId": "507ffbd3-0f6e-4aaf-ece4-6edb84ebb5d0"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 2500x1200 with 5 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_compare_distributions(train_labels, test_labels, title1='train labels', title2='test labels')"
      ],
      "metadata": {
        "id": "u-e1P9AJ7H5r",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        },
        "outputId": "542e05f2-24a5-4dfa-a064-e36f6b78484f"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "class_weight = dict(zip(range(0, 7), (((data2[data2[' Usage']=='Training']['emotion'].value_counts()).sort_index())/len(data2[data2[' Usage']=='Training']['emotion'])).tolist()))"
      ],
      "metadata": {
        "id": "6IkqiKEY7XDQ"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "class_weight"
      ],
      "metadata": {
        "id": "XHGRDtp57XLu",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e8faacc0-7728-4b98-9c18-965a4456eee7"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{0: 0.1275339185953711,\n",
              " 1: 0.09743016759776536,\n",
              " 2: 0.13079010375099762,\n",
              " 3: 0.2303272146847566,\n",
              " 4: 0.15418994413407822,\n",
              " 5: 0.1012290502793296,\n",
              " 6: 0.15849960095770152}"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model = models.Sequential()\n",
        "\n",
        "model.add(Conv2D(64, (3, 3), activation='relu', input_shape=(48, 48, 1)))\n",
        "model.add(MaxPool2D((2, 2)))\n",
        "model.add(BatchNormalization())\n",
        "\n",
        "model.add(Conv2D(128, (3, 3), activation='relu'))\n",
        "model.add(MaxPool2D((2, 2)))\n",
        "model.add(BatchNormalization())\n",
        "\n",
        "model.add(Conv2D(256, (3, 3), activation='relu'))\n",
        "model.add(MaxPool2D((2, 2)))\n",
        "model.add(BatchNormalization())\n",
        "\n",
        "model.add(Flatten())\n",
        "\n",
        "model.add(Dense(256, activation='relu'))\n",
        "model.add(Dense(7, activation='softmax'))"
      ],
      "metadata": {
        "id": "qjPLgQhc7XRO"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.optimizers.optimizer import learning_rate_schedule\n",
        "model.compile(optimizer=Adam(learning_rate=0.0001), loss='categorical_crossentropy', metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "CKzMS45u7XVv"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "id": "FfB9HI_d7mVb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e724331e-e92a-40f5-96c1-d50cd5b4d6a9"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d (Conv2D)             (None, 46, 46, 64)        640       \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 23, 23, 64)       0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 23, 23, 64)       256       \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 21, 21, 128)       73856     \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 10, 10, 128)      0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 10, 10, 128)      512       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " conv2d_2 (Conv2D)           (None, 8, 8, 256)         295168    \n",
            "                                                                 \n",
            " max_pooling2d_2 (MaxPooling  (None, 4, 4, 256)        0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " batch_normalization_2 (Batc  (None, 4, 4, 256)        1024      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 4096)              0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 256)               1048832   \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 7)                 1799      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 1,422,087\n",
            "Trainable params: 1,421,191\n",
            "Non-trainable params: 896\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history = model.fit(train_images, train_labels,\n",
        "                    validation_data = (val_images, val_labels),\n",
        "                    class_weight = class_weight,\n",
        "                    epochs=10,\n",
        "                    batch_size=25)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_dFlm6KU8Nep",
        "outputId": "5dc72766-2108-4085-be71-25e9a3f6fa7d"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "1253/1253 [==============================] - 20s 7ms/step - loss: 0.2333 - accuracy: 0.4173 - val_loss: 1.4110 - val_accuracy: 0.4700\n",
            "Epoch 2/10\n",
            "1253/1253 [==============================] - 8s 7ms/step - loss: 0.1648 - accuracy: 0.5867 - val_loss: 1.3635 - val_accuracy: 0.5182\n",
            "Epoch 3/10\n",
            "1253/1253 [==============================] - 8s 6ms/step - loss: 0.1275 - accuracy: 0.6842 - val_loss: 1.3016 - val_accuracy: 0.5361\n",
            "Epoch 4/10\n",
            "1253/1253 [==============================] - 10s 8ms/step - loss: 0.0972 - accuracy: 0.7684 - val_loss: 1.3388 - val_accuracy: 0.5438\n",
            "Epoch 5/10\n",
            "1253/1253 [==============================] - 8s 7ms/step - loss: 0.0703 - accuracy: 0.8384 - val_loss: 1.3816 - val_accuracy: 0.5486\n",
            "Epoch 6/10\n",
            "1253/1253 [==============================] - 7s 6ms/step - loss: 0.0479 - accuracy: 0.8987 - val_loss: 1.5140 - val_accuracy: 0.5560\n",
            "Epoch 7/10\n",
            "1253/1253 [==============================] - 8s 7ms/step - loss: 0.0314 - accuracy: 0.9409 - val_loss: 1.5420 - val_accuracy: 0.5512\n",
            "Epoch 8/10\n",
            "1253/1253 [==============================] - 8s 6ms/step - loss: 0.0213 - accuracy: 0.9649 - val_loss: 1.6835 - val_accuracy: 0.5407\n",
            "Epoch 9/10\n",
            "1253/1253 [==============================] - 9s 7ms/step - loss: 0.0157 - accuracy: 0.9764 - val_loss: 1.7972 - val_accuracy: 0.5481\n",
            "Epoch 10/10\n",
            "1253/1253 [==============================] - 9s 7ms/step - loss: 0.0128 - accuracy: 0.9815 - val_loss: 1.9122 - val_accuracy: 0.5484\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_loss, test_acc = model.evaluate(test_images, test_labels)\n",
        "print('test accuracy:', test_acc)"
      ],
      "metadata": {
        "id": "gPD0Wfk18QQw",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ccd79120-9571-43da-aa9b-d44de52644b7"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "123/123 [==============================] - 1s 6ms/step - loss: 2.0348 - accuracy: 0.5355\n",
            "test accuracy: 0.5355414152145386\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pred_test_labels = model.predict(test_images)"
      ],
      "metadata": {
        "id": "DtcwK-IS8QT9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "df05ea41-7ed5-4de9-9e59-e6041323b1ec"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "123/123 [==============================] - 1s 3ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pred_test_labels = np.argmax(pred_test_labels, axis=1)  # if test_labels contains probabilities"
      ],
      "metadata": {
        "id": "4zP9QoB9vOib"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "pred_test_labels"
      ],
      "metadata": {
        "id": "IiVV3XwYvlsg",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "14f84816-2c57-4fd2-fcd7-8138a6dfaae0"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([5, 1, 0, ..., 2, 1, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_labels = np.argmax(test_labels, axis=1)"
      ],
      "metadata": {
        "id": "Co51wyFuv_tV"
      },
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_labels"
      ],
      "metadata": {
        "id": "u6o2u5rmwSwY",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a64702e4-ed69-44c8-b949-f810c5abde44"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 1, 4, ..., 1, 1, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "loss = history.history['loss']\n",
        "loss_val = history.history['val_loss']\n",
        "epochs = range(1, len(loss)+1)\n",
        "plt.plot(epochs, loss, 'bo', label='loss_train')\n",
        "plt.plot(epochs, loss_val, 'b', label='loss_val')\n",
        "plt.title('value of the loss function')\n",
        "plt.xlabel('epochs')\n",
        "plt.ylabel('value of the loss function')\n",
        "plt.legend()\n",
        "plt.grid()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "kHVcvuMO8QYo",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "outputId": "00cac64c-8831-4d73-8ca4-7fb23a640acc"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "acc = history.history['accuracy']\n",
        "acc_val = history.history['val_accuracy']\n",
        "epochs = range(1, len(loss)+1)\n",
        "plt.plot(epochs, acc, 'bo', label='accuracy_train')\n",
        "plt.plot(epochs, acc_val, 'b', label='accuracy_val')\n",
        "plt.title('accuracy')\n",
        "plt.xlabel('epochs')\n",
        "plt.ylabel('value of accuracy')\n",
        "plt.legend()\n",
        "plt.grid()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "-3XKsbPU8aYl",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "outputId": "ead35b84-8cdc-414f-c45e-28f1f2c2fd89"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cm = confusion_matrix(test_labels, pred_test_labels)"
      ],
      "metadata": {
        "id": "ry_YOEBaucyb"
      },
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "cm"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-TxnLg7Z9lGp",
        "outputId": "58fac26b-d240-4f62-d4d9-f659d04dcc3f"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[210,   3,  62,  31,  89,  19,  53],\n",
              "       [105, 182,  35,  14,  42,   0,  14],\n",
              "       [ 55,   2, 191,  28, 107,  54,  59],\n",
              "       [ 43,   2,  46, 610,  75,  37,  82],\n",
              "       [ 69,   4,  88,  39, 315,  31, 107],\n",
              "       [ 17,   0,  47,  20,  11, 308,  12],\n",
              "       [ 57,   1,  66,  54, 118,  25, 286]])"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "alexnet = models.Sequential()\n",
        "\n",
        "# Layer 1\n",
        "alexnet.add(Conv2D(96, (11, 11), input_shape=(48, 48, 1), padding='same', activation='tanh'))\n",
        "alexnet.add(BatchNormalization())\n",
        "alexnet.add(MaxPool2D(pool_size=(2, 2)))\n",
        "\n",
        "# Layer 2\n",
        "alexnet.add(Conv2D(256, (5, 5), padding='same', activation='tanh'))\n",
        "alexnet.add(BatchNormalization())\n",
        "alexnet.add(MaxPool2D(pool_size=(2, 2)))\n",
        "\n",
        "# Layer 3\n",
        "alexnet.add(ZeroPadding2D((1, 1)))\n",
        "alexnet.add(Conv2D(512, (3, 3), padding='same', activation='tanh'))\n",
        "alexnet.add(BatchNormalization())\n",
        "alexnet.add(MaxPool2D(pool_size=(2, 2)))\n",
        "\n",
        "# Layer 4\n",
        "alexnet.add(ZeroPadding2D((1, 1)))\n",
        "alexnet.add(Conv2D(1024, (3, 3), padding='same', activation='tanh'))\n",
        "alexnet.add(BatchNormalization())\n",
        "\n",
        "# Layer 5\n",
        "alexnet.add(ZeroPadding2D((1, 1)))\n",
        "alexnet.add(Conv2D(1024, (3, 3), padding='same', activation='tanh'))\n",
        "alexnet.add(BatchNormalization())\n",
        "alexnet.add(MaxPool2D(pool_size=(2, 2)))\n",
        "\n",
        "# Layer 6\n",
        "alexnet.add(Flatten())\n",
        "alexnet.add(Dense(4096, activation='tanh'))\n",
        "alexnet.add(BatchNormalization())\n",
        "alexnet.add(Dropout(0.5))\n",
        "\n",
        "# Layer 7\n",
        "alexnet.add(Dense(4096, activation='tanh'))\n",
        "alexnet.add(BatchNormalization())\n",
        "alexnet.add(Dropout(0.5))\n",
        "\n",
        "# Layer 8\n",
        "alexnet.add(Dense(7, activation='softmax'))\n",
        "alexnet.add(BatchNormalization())\n",
        "\n",
        "# print model summary\n",
        "alexnet.summary()"
      ],
      "metadata": {
        "id": "Bx5eq7LacvIp",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "595a81e1-63f6-4186-ae32-b0c5350b6d52"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_3 (Conv2D)           (None, 48, 48, 96)        11712     \n",
            "                                                                 \n",
            " batch_normalization_3 (Batc  (None, 48, 48, 96)       384       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " max_pooling2d_3 (MaxPooling  (None, 24, 24, 96)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_4 (Conv2D)           (None, 24, 24, 256)       614656    \n",
            "                                                                 \n",
            " batch_normalization_4 (Batc  (None, 24, 24, 256)      1024      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " max_pooling2d_4 (MaxPooling  (None, 12, 12, 256)      0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " zero_padding2d (ZeroPadding  (None, 14, 14, 256)      0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_5 (Conv2D)           (None, 14, 14, 512)       1180160   \n",
            "                                                                 \n",
            " batch_normalization_5 (Batc  (None, 14, 14, 512)      2048      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " max_pooling2d_5 (MaxPooling  (None, 7, 7, 512)        0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " zero_padding2d_1 (ZeroPaddi  (None, 9, 9, 512)        0         \n",
            " ng2D)                                                           \n",
            "                                                                 \n",
            " conv2d_6 (Conv2D)           (None, 9, 9, 1024)        4719616   \n",
            "                                                                 \n",
            " batch_normalization_6 (Batc  (None, 9, 9, 1024)       4096      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " zero_padding2d_2 (ZeroPaddi  (None, 11, 11, 1024)     0         \n",
            " ng2D)                                                           \n",
            "                                                                 \n",
            " conv2d_7 (Conv2D)           (None, 11, 11, 1024)      9438208   \n",
            "                                                                 \n",
            " batch_normalization_7 (Batc  (None, 11, 11, 1024)     4096      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " max_pooling2d_6 (MaxPooling  (None, 5, 5, 1024)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten_1 (Flatten)         (None, 25600)             0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 4096)              104861696 \n",
            "                                                                 \n",
            " batch_normalization_8 (Batc  (None, 4096)             16384     \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 4096)              0         \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 4096)              16781312  \n",
            "                                                                 \n",
            " batch_normalization_9 (Batc  (None, 4096)             16384     \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 4096)              0         \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 7)                 28679     \n",
            "                                                                 \n",
            " batch_normalization_10 (Bat  (None, 7)                28        \n",
            " chNormalization)                                                \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 137,680,483\n",
            "Trainable params: 137,658,261\n",
            "Non-trainable params: 22,222\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "alexnet.compile(optimizer=Adam(learning_rate=0.0001), loss='categorical_crossentropy', metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "yQTBz-TSg49X"
      },
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "history = alexnet.fit(train_images, train_labels,\n",
        "                    validation_data = (val_images, val_labels),\n",
        "                    class_weight = class_weight,\n",
        "                    epochs=25,\n",
        "                    batch_size=50)"
      ],
      "metadata": {
        "id": "Qzhgj_96h8_6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "41c93c8e-c4a3-40e3-feec-c2746026f110"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/25\n",
            "627/627 [==============================] - 95s 135ms/step - loss: 1.1558 - accuracy: 0.1627 - val_loss: 10.1853 - val_accuracy: 0.1245\n",
            "Epoch 2/25\n",
            "627/627 [==============================] - 83s 132ms/step - loss: 1.1206 - accuracy: 0.1798 - val_loss: 7.3464 - val_accuracy: 0.1653\n",
            "Epoch 3/25\n",
            "627/627 [==============================] - 83s 132ms/step - loss: 1.0843 - accuracy: 0.1511 - val_loss: 5.6462 - val_accuracy: 0.1485\n",
            "Epoch 4/25\n",
            "627/627 [==============================] - 83s 132ms/step - loss: 0.8785 - accuracy: 0.1207 - val_loss: 8.2915 - val_accuracy: 0.1306\n",
            "Epoch 5/25\n",
            "627/627 [==============================] - 83s 132ms/step - loss: 1.0347 - accuracy: 0.1516 - val_loss: 5.8404 - val_accuracy: 0.1595\n",
            "Epoch 6/25\n",
            "627/627 [==============================] - 83s 132ms/step - loss: 1.1863 - accuracy: 0.1708 - val_loss: 7.4764 - val_accuracy: 0.1429\n",
            "Epoch 7/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 1.4858 - accuracy: 0.1715 - val_loss: 11.7959 - val_accuracy: 0.1592\n",
            "Epoch 8/25\n",
            "627/627 [==============================] - 80s 128ms/step - loss: 1.4328 - accuracy: 0.1747 - val_loss: 10.5428 - val_accuracy: 0.1595\n",
            "Epoch 9/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 1.1949 - accuracy: 0.1947 - val_loss: 6.3681 - val_accuracy: 0.1781\n",
            "Epoch 10/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 1.1577 - accuracy: 0.2068 - val_loss: 6.5606 - val_accuracy: 0.1998\n",
            "Epoch 11/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 0.7711 - accuracy: 0.2109 - val_loss: 3.9275 - val_accuracy: 0.2029\n",
            "Epoch 12/25\n",
            "627/627 [==============================] - 80s 127ms/step - loss: 1.0601 - accuracy: 0.1860 - val_loss: 7.0974 - val_accuracy: 0.1559\n",
            "Epoch 13/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 1.2310 - accuracy: 0.1579 - val_loss: 6.1939 - val_accuracy: 0.1521\n",
            "Epoch 14/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 1.2403 - accuracy: 0.1860 - val_loss: 9.5312 - val_accuracy: 0.2059\n",
            "Epoch 15/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 1.1918 - accuracy: 0.1935 - val_loss: 7.8980 - val_accuracy: 0.2006\n",
            "Epoch 16/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 1.0675 - accuracy: 0.1944 - val_loss: 7.5186 - val_accuracy: 0.1712\n",
            "Epoch 17/25\n",
            "627/627 [==============================] - 80s 128ms/step - loss: 1.1016 - accuracy: 0.1757 - val_loss: 6.2051 - val_accuracy: 0.1771\n",
            "Epoch 18/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 1.1840 - accuracy: 0.1757 - val_loss: 5.9969 - val_accuracy: 0.1585\n",
            "Epoch 19/25\n",
            "627/627 [==============================] - 80s 127ms/step - loss: 1.2161 - accuracy: 0.1779 - val_loss: 8.0995 - val_accuracy: 0.1725\n",
            "Epoch 20/25\n",
            "627/627 [==============================] - 80s 128ms/step - loss: 1.3151 - accuracy: 0.1872 - val_loss: 7.7778 - val_accuracy: 0.1832\n",
            "Epoch 21/25\n",
            "627/627 [==============================] - 80s 127ms/step - loss: 1.3155 - accuracy: 0.1831 - val_loss: 9.2773 - val_accuracy: 0.1796\n",
            "Epoch 22/25\n",
            "627/627 [==============================] - 80s 127ms/step - loss: 1.2487 - accuracy: 0.1909 - val_loss: 7.0358 - val_accuracy: 0.1919\n",
            "Epoch 23/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 1.1797 - accuracy: 0.2034 - val_loss: 7.6301 - val_accuracy: 0.2013\n",
            "Epoch 24/25\n",
            "627/627 [==============================] - 80s 127ms/step - loss: 1.2004 - accuracy: 0.1994 - val_loss: 7.1335 - val_accuracy: 0.1911\n",
            "Epoch 25/25\n",
            "627/627 [==============================] - 82s 131ms/step - loss: 1.1892 - accuracy: 0.1987 - val_loss: 6.4978 - val_accuracy: 0.1932\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.save('/content/drive/MyDrive/Mood-Detection-Project/challenges-in-representation-learning-facial-expression-recognition-challenge/model.h5')\n"
      ],
      "metadata": {
        "id": "MJh1zrAk-4Ee"
      },
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import cv2\n",
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "\n",
        "# Load pre-trained Haar Cascade classifier for face detection\n",
        "face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')\n",
        "\n",
        "# Load pre-trained machine learning model for mood prediction\n",
        "model = tf.keras.models.load_model('/content/drive/MyDrive/Mood-Detection-Project/challenges-in-representation-learning-facial-expression-recognition-challenge/model.h5')\n",
        "\n",
        "# Define a dictionary to map predicted class indices to mood labels\n",
        "mood_labels = {0: 'angry', 1: 'disgust', 2: 'fear', 3: 'happy', 4: 'sad', 5: 'surprise', 6: 'neutral'}\n",
        "\n",
        "# Open video capture device\n",
        "cap = cv2.VideoCapture(0)\n",
        "\n",
        "cap.isOpened()\n",
        "\n",
        "# Loop over frames from video capture device\n",
        "while True:\n",
        "    # Read frame from video capture device\n",
        "    ret, frame = cap.read()\n",
        "\n",
        "    # Convert frame to grayscale\n",
        "    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)\n",
        "\n",
        "    # Detect faces in frame using Haar Cascade classifier\n",
        "    faces = face_cascade.detectMultiScale(gray, scaleFactor=1.1, minNeighbors=5)\n",
        "\n",
        "    # Loop over detected faces\n",
        "    for (x, y, w, h) in faces:\n",
        "        # Extract face ROI (Region of Interest)\n",
        "        face_roi = gray[y:y+h, x:x+w]\n",
        "        # Resize face ROI to fit model input size\n",
        "        face_roi = cv2.resize(face_roi, (48, 48))\n",
        "        # Normalize pixel values to range [0, 1]\n",
        "        face_roi = face_roi / 255.0\n",
        "        # Reshape face ROI to match model input shape\n",
        "        face_roi = np.reshape(face_roi, (1, 48, 48, 1))\n",
        "        # Make mood prediction using pre-trained model\n",
        "        prediction = model.predict(face_roi)\n",
        "        # Get predicted mood label\n",
        "        mood_label = mood_labels[np.argmax(prediction)]\n",
        "        # Draw bounding box around detected face\n",
        "        cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 0), 2)\n",
        "        # Write predicted mood label on top of bounding box\n",
        "        cv2.putText(frame, mood_label, (x, y-10), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (0, 255, 0), 2)\n",
        "\n",
        "    # Display output frame\n",
        "    cv2.imshow('Output', frame)\n",
        "\n",
        "    # Exit loop if 'q' key is pressed\n",
        "    if cv2.waitKey(1) == ord('q'):\n",
        "        break\n",
        "\n",
        "# Release video capture device and close all windows\n",
        "cap.release()\n",
        "cv2.destroyAllWindows()"
      ],
      "metadata": {
        "id": "KroNn0Yjwl3N",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "92f4202c-f551-4694-a44b-c736ef12702b"
      },
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "False"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    }
  ]
}