SAVEE emotion recognition.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "V1+A4.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I0iQ9jS8nZ6Z"
      },
      "source": [
        "Dataset loading and preparing\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c2H--q_sefN5",
        "outputId": "013146de-90d4-45a3-8d09-58ba589f58a6"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "26Q6HLq6naNY"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import tensorflow as tf\n",
        "import keras\n",
        "import matplotlib.pyplot as plt\n",
        "from matplotlib import cm\n",
        "import random\n",
        "import time\n",
        "%matplotlib inline"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ytffdZ3sw-VF"
      },
      "source": [
        "data= np.load('/content/drive/MyDrive/SAVEE fusion labelled data.npy', allow_pickle=True)"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GKlXjgnBw-eX"
      },
      "source": [
        "data_dict = data.item()\n",
        "video_features = data_dict['video_features']\n",
        "audio_features = data_dict['audio_features']\n",
        "labels = data_dict['labels']"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MzlAXa_WxUKm",
        "outputId": "7a04bed8-138d-4296-cb20-83c2d637a084"
      },
      "source": [
        "print(video_features.shape)\n",
        "print(audio_features.shape)\n",
        "print(labels.shape)"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(480, 6, 64, 64, 3)\n",
            "(480, 181)\n",
            "(480,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nYn-tUu_xagw",
        "outputId": "eed0888b-c89a-419f-ab1c-e81515f979ac"
      },
      "source": [
        "from sklearn.preprocessing import LabelEncoder\n",
        "from tensorflow.keras.utils import to_categorical\n",
        "lb = LabelEncoder()\n",
        "labels = to_categorical(lb.fit_transform(labels))\n",
        "print(labels.shape)"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(480, 7)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uZqY_tbfyu1r",
        "outputId": "7896c128-4e99-4119-9121-1b72acd5ad05"
      },
      "source": [
        "audio_features=audio_features.reshape(480,181,1)\n",
        "print(\"Audio representation shape:\",audio_features.shape)"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Audio representation shape: (480, 181, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sVnSJEHn_JH7"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "video_features_train, video_features_val,audio_features_train, audio_features_val,labels_train, labels_val = train_test_split(video_features,audio_features,labels, test_size=0.2,random_state=0)\n",
        "video_features_val, video_features_test,audio_features_val, audio_features_test,labels_val, labels_test = train_test_split(video_features_val,audio_features_val,labels_val, test_size=0.5,random_state=0)"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "P1SW5n0m_a5t",
        "outputId": "acffd6bf-ac85-452d-d83e-4f26844fc2a6"
      },
      "source": [
        "print(video_features_train.shape)\n",
        "print(video_features_val.shape)\n",
        "print(video_features_test.shape)\n",
        "print(audio_features_train.shape)\n",
        "print(audio_features_val.shape)\n",
        "print(audio_features_test.shape)\n",
        "print(labels_train.shape)\n",
        "print(labels_val.shape)\n",
        "print(labels_test.shape)"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(384, 6, 64, 64, 3)\n",
            "(48, 6, 64, 64, 3)\n",
            "(48, 6, 64, 64, 3)\n",
            "(384, 181, 1)\n",
            "(48, 181, 1)\n",
            "(48, 181, 1)\n",
            "(384, 7)\n",
            "(48, 7)\n",
            "(48, 7)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XRyROLuonJz7"
      },
      "source": [
        "Video model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CpCx0ToZzvup"
      },
      "source": [
        "from tensorflow.keras.models import *\n",
        "from tensorflow.keras.layers import *\n",
        "from tensorflow.keras.callbacks import *\n",
        "from tensorflow.keras.optimizers import *"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lDUPXZQRje_d"
      },
      "source": [
        "input_video = Input(shape=(6,64,64,3))\n",
        "    \n",
        "conv = ConvLSTM2D(64, 3, activation='relu', padding = 'same')(input_video)\n",
        "conv= Convolution2D(64, 3, activation='relu', padding = 'same')(conv)\n",
        "pool = MaxPooling2D((2, 2), strides=(2, 2))(conv)\n",
        "    \n",
        "conv = Convolution2D(128, 3, activation='relu', padding = 'same')(pool)\n",
        "pool = MaxPooling2D((2, 2), strides=(2, 2))(conv)\n",
        "conv = Convolution2D(128, 3, activation='relu', padding = 'same')(pool)\n",
        "pool = MaxPooling2D((2, 2), strides=(2, 2))(conv)\n",
        "    \n",
        "conv = Convolution2D(256, 3, activation='relu', padding = 'same')(pool)\n",
        "#conv3 = Convolution2D(256, 3, activation='relu', padding = 'same')(conv3)\n",
        "#conv3 = Convolution2D(256, 3, activation='relu', padding = 'same')(conv3)\n",
        "pool = MaxPooling2D((2, 2), strides=(2, 2))(conv)\n",
        "\n",
        "conv = Convolution2D(512, 3, activation='relu', padding = 'same')(pool)\n",
        "#conv4 = Convolution2D(512, 3, activation='relu', padding = 'same')(conv4)\n",
        "#conv4 = Convolution2D(512, 3, activation='relu', padding = 'same')(conv4)\n",
        "pool = MaxPooling2D((2, 2), strides=(2, 2))(conv)\n",
        "    \n",
        "output_video = Flatten()(pool)"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wWODmniZqLwM"
      },
      "source": [
        "Audio model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wT_fTNWWjiXQ"
      },
      "source": [
        "from tensorflow.keras.regularizers import l2\n",
        "input_audio = Input(shape=(181,1))\n",
        "    \n",
        "conv = Convolution1D(64, 10, activation='relu')(input_audio)\n",
        "conv = Convolution1D(64, 10, activation='relu')(conv)\n",
        "pool = MaxPooling1D((8))(conv)\n",
        "conv = Convolution1D(128, 10, activation='relu',kernel_regularizer=l2(0.01), bias_regularizer=l2(0.01))(pool)\n",
        "pool = MaxPooling1D((8))(conv)\n",
        "drop = Dropout(0.4)(pool)\n",
        "    \n",
        "output_audio = Flatten()(drop)"
      ],
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gFJGyrnMqWgR"
      },
      "source": [
        "Concatenation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "G-1-bySRjlX2",
        "outputId": "f59f0e04-d7e6-470a-eedd-d10a4cedbdda"
      },
      "source": [
        "#Now concatenate all the layers to make a multimodal\n",
        "concatenate = keras.layers.Concatenate()([output_video, output_audio])\n",
        "final_layer = Dense(128, activation='relu')(concatenate)\n",
        "y_predict_emotions = Dense(7, activation='softmax')(final_layer)\n",
        "model = Model(inputs=[input_video, input_audio], outputs=[y_predict_emotions])\n",
        "tf.keras.utils.plot_model(model, to_file='multimodal model.png', show_shapes=True, show_layer_names=True)"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oxwJ5PoLjeyV",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1dbc4eb5-05dd-4a8d-9edf-f0d2e46291a8"
      },
      "source": [
        "opt = tf.keras.optimizers.Adam(learning_rate=0.001)\n",
        "model.compile(optimizer = opt , loss = 'categorical_crossentropy', metrics = ['accuracy'])\n",
        "start=time.time()\n",
        "history = model.fit([video_features_train,audio_features_train],labels_train,\n",
        "                    epochs = 120, \n",
        "                    batch_size=32,\n",
        "                    validation_data=([video_features_val,audio_features_val],labels_val))\n",
        "end=time.time()\n",
        "train_time_avg=(end-start)/120"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/120\n",
            "12/12 [==============================] - 12s 784ms/step - loss: 2.6473 - accuracy: 0.3151 - val_loss: 2.3531 - val_accuracy: 0.3333\n",
            "Epoch 2/120\n",
            "12/12 [==============================] - 9s 721ms/step - loss: 2.2515 - accuracy: 0.3594 - val_loss: 2.1314 - val_accuracy: 0.4167\n",
            "Epoch 3/120\n",
            "12/12 [==============================] - 9s 716ms/step - loss: 1.9934 - accuracy: 0.3698 - val_loss: 1.9402 - val_accuracy: 0.4167\n",
            "Epoch 4/120\n",
            "12/12 [==============================] - 9s 718ms/step - loss: 1.8427 - accuracy: 0.3932 - val_loss: 1.7285 - val_accuracy: 0.4375\n",
            "Epoch 5/120\n",
            "12/12 [==============================] - 9s 715ms/step - loss: 1.7117 - accuracy: 0.4271 - val_loss: 1.6249 - val_accuracy: 0.4583\n",
            "Epoch 6/120\n",
            "12/12 [==============================] - 9s 714ms/step - loss: 1.5501 - accuracy: 0.5078 - val_loss: 1.6828 - val_accuracy: 0.3958\n",
            "Epoch 7/120\n",
            "12/12 [==============================] - 9s 713ms/step - loss: 1.4852 - accuracy: 0.4948 - val_loss: 1.5710 - val_accuracy: 0.5000\n",
            "Epoch 8/120\n",
            "12/12 [==============================] - 9s 715ms/step - loss: 1.4282 - accuracy: 0.5000 - val_loss: 1.6224 - val_accuracy: 0.4583\n",
            "Epoch 9/120\n",
            "12/12 [==============================] - 9s 713ms/step - loss: 1.4127 - accuracy: 0.5339 - val_loss: 1.6848 - val_accuracy: 0.4167\n",
            "Epoch 10/120\n",
            "12/12 [==============================] - 8s 711ms/step - loss: 1.3096 - accuracy: 0.5573 - val_loss: 1.5739 - val_accuracy: 0.4583\n",
            "Epoch 11/120\n",
            "12/12 [==============================] - 9s 711ms/step - loss: 1.1573 - accuracy: 0.6198 - val_loss: 1.3886 - val_accuracy: 0.5000\n",
            "Epoch 12/120\n",
            "12/12 [==============================] - 8s 711ms/step - loss: 1.1404 - accuracy: 0.6094 - val_loss: 1.3240 - val_accuracy: 0.5208\n",
            "Epoch 13/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.9927 - accuracy: 0.6667 - val_loss: 1.0506 - val_accuracy: 0.6042\n",
            "Epoch 14/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.9121 - accuracy: 0.6875 - val_loss: 1.1362 - val_accuracy: 0.5625\n",
            "Epoch 15/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.8577 - accuracy: 0.7708 - val_loss: 0.8353 - val_accuracy: 0.7708\n",
            "Epoch 16/120\n",
            "12/12 [==============================] - 8s 711ms/step - loss: 0.7634 - accuracy: 0.7734 - val_loss: 0.8316 - val_accuracy: 0.7083\n",
            "Epoch 17/120\n",
            "12/12 [==============================] - 8s 711ms/step - loss: 0.6480 - accuracy: 0.8047 - val_loss: 0.7959 - val_accuracy: 0.7292\n",
            "Epoch 18/120\n",
            "12/12 [==============================] - 9s 711ms/step - loss: 0.6051 - accuracy: 0.8333 - val_loss: 1.0966 - val_accuracy: 0.6042\n",
            "Epoch 19/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.6015 - accuracy: 0.8359 - val_loss: 0.9834 - val_accuracy: 0.6458\n",
            "Epoch 20/120\n",
            "12/12 [==============================] - 8s 710ms/step - loss: 0.5793 - accuracy: 0.8359 - val_loss: 0.8517 - val_accuracy: 0.6875\n",
            "Epoch 21/120\n",
            "12/12 [==============================] - 9s 713ms/step - loss: 0.5085 - accuracy: 0.8672 - val_loss: 0.6286 - val_accuracy: 0.7708\n",
            "Epoch 22/120\n",
            "12/12 [==============================] - 9s 711ms/step - loss: 0.4606 - accuracy: 0.8932 - val_loss: 0.5928 - val_accuracy: 0.8750\n",
            "Epoch 23/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.4256 - accuracy: 0.8932 - val_loss: 0.4752 - val_accuracy: 0.8125\n",
            "Epoch 24/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.4222 - accuracy: 0.8958 - val_loss: 0.5726 - val_accuracy: 0.8333\n",
            "Epoch 25/120\n",
            "12/12 [==============================] - 9s 733ms/step - loss: 0.3604 - accuracy: 0.9245 - val_loss: 0.5834 - val_accuracy: 0.7500\n",
            "Epoch 26/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.2825 - accuracy: 0.9479 - val_loss: 0.5394 - val_accuracy: 0.8125\n",
            "Epoch 27/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.2744 - accuracy: 0.9531 - val_loss: 0.4177 - val_accuracy: 0.8333\n",
            "Epoch 28/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.1980 - accuracy: 0.9766 - val_loss: 0.3656 - val_accuracy: 0.8958\n",
            "Epoch 29/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.1744 - accuracy: 0.9818 - val_loss: 0.3330 - val_accuracy: 0.8958\n",
            "Epoch 30/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.1672 - accuracy: 0.9792 - val_loss: 0.6176 - val_accuracy: 0.8125\n",
            "Epoch 31/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.1369 - accuracy: 0.9896 - val_loss: 0.4917 - val_accuracy: 0.8542\n",
            "Epoch 32/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.1264 - accuracy: 0.9922 - val_loss: 0.4475 - val_accuracy: 0.8542\n",
            "Epoch 33/120\n",
            "12/12 [==============================] - 8s 710ms/step - loss: 0.1927 - accuracy: 0.9453 - val_loss: 0.4779 - val_accuracy: 0.9167\n",
            "Epoch 34/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.1712 - accuracy: 0.9661 - val_loss: 0.3928 - val_accuracy: 0.8333\n",
            "Epoch 35/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.1803 - accuracy: 0.9661 - val_loss: 1.1752 - val_accuracy: 0.6875\n",
            "Epoch 36/120\n",
            "12/12 [==============================] - 8s 711ms/step - loss: 0.2046 - accuracy: 0.9661 - val_loss: 0.4462 - val_accuracy: 0.8542\n",
            "Epoch 37/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.1388 - accuracy: 0.9844 - val_loss: 0.4376 - val_accuracy: 0.8333\n",
            "Epoch 38/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.1362 - accuracy: 0.9792 - val_loss: 0.5068 - val_accuracy: 0.8333\n",
            "Epoch 39/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.1449 - accuracy: 0.9844 - val_loss: 0.3605 - val_accuracy: 0.8542\n",
            "Epoch 40/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.1279 - accuracy: 0.9844 - val_loss: 0.5233 - val_accuracy: 0.8125\n",
            "Epoch 41/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.2102 - accuracy: 0.9609 - val_loss: 0.5891 - val_accuracy: 0.7500\n",
            "Epoch 42/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.3347 - accuracy: 0.9141 - val_loss: 0.5282 - val_accuracy: 0.8750\n",
            "Epoch 43/120\n",
            "12/12 [==============================] - 8s 710ms/step - loss: 0.2286 - accuracy: 0.9453 - val_loss: 0.3922 - val_accuracy: 0.8333\n",
            "Epoch 44/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.1786 - accuracy: 0.9583 - val_loss: 0.3589 - val_accuracy: 0.8750\n",
            "Epoch 45/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.1115 - accuracy: 0.9974 - val_loss: 0.4115 - val_accuracy: 0.8958\n",
            "Epoch 46/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0878 - accuracy: 1.0000 - val_loss: 0.3190 - val_accuracy: 0.8958\n",
            "Epoch 47/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0911 - accuracy: 0.9948 - val_loss: 0.4368 - val_accuracy: 0.8958\n",
            "Epoch 48/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0834 - accuracy: 0.9922 - val_loss: 0.2344 - val_accuracy: 0.9167\n",
            "Epoch 49/120\n",
            "12/12 [==============================] - 8s 705ms/step - loss: 0.0811 - accuracy: 0.9974 - val_loss: 0.3513 - val_accuracy: 0.9167\n",
            "Epoch 50/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0793 - accuracy: 0.9948 - val_loss: 0.2828 - val_accuracy: 0.8958\n",
            "Epoch 51/120\n",
            "12/12 [==============================] - 8s 703ms/step - loss: 0.0828 - accuracy: 0.9922 - val_loss: 0.4103 - val_accuracy: 0.8958\n",
            "Epoch 52/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.1286 - accuracy: 0.9714 - val_loss: 0.4372 - val_accuracy: 0.8750\n",
            "Epoch 53/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.0949 - accuracy: 0.9896 - val_loss: 0.2341 - val_accuracy: 0.9375\n",
            "Epoch 54/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0929 - accuracy: 0.9870 - val_loss: 0.2168 - val_accuracy: 0.9583\n",
            "Epoch 55/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.0976 - accuracy: 0.9844 - val_loss: 0.4222 - val_accuracy: 0.8750\n",
            "Epoch 56/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0691 - accuracy: 0.9948 - val_loss: 0.2681 - val_accuracy: 0.9167\n",
            "Epoch 57/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0819 - accuracy: 0.9870 - val_loss: 0.2718 - val_accuracy: 0.9375\n",
            "Epoch 58/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0562 - accuracy: 1.0000 - val_loss: 0.2894 - val_accuracy: 0.9167\n",
            "Epoch 59/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0506 - accuracy: 1.0000 - val_loss: 0.2087 - val_accuracy: 0.9583\n",
            "Epoch 60/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0460 - accuracy: 1.0000 - val_loss: 0.1996 - val_accuracy: 0.9583\n",
            "Epoch 61/120\n",
            "12/12 [==============================] - 8s 705ms/step - loss: 0.0464 - accuracy: 1.0000 - val_loss: 0.1728 - val_accuracy: 0.9375\n",
            "Epoch 62/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0519 - accuracy: 0.9948 - val_loss: 0.2911 - val_accuracy: 0.8958\n",
            "Epoch 63/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0527 - accuracy: 0.9974 - val_loss: 0.2439 - val_accuracy: 0.9167\n",
            "Epoch 64/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0443 - accuracy: 1.0000 - val_loss: 0.1647 - val_accuracy: 0.9583\n",
            "Epoch 65/120\n",
            "12/12 [==============================] - 8s 710ms/step - loss: 0.0391 - accuracy: 1.0000 - val_loss: 0.1737 - val_accuracy: 0.9583\n",
            "Epoch 66/120\n",
            "12/12 [==============================] - 8s 705ms/step - loss: 0.0373 - accuracy: 1.0000 - val_loss: 0.1413 - val_accuracy: 0.9583\n",
            "Epoch 67/120\n",
            "12/12 [==============================] - 8s 710ms/step - loss: 0.0354 - accuracy: 1.0000 - val_loss: 0.1788 - val_accuracy: 0.9375\n",
            "Epoch 68/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0337 - accuracy: 1.0000 - val_loss: 0.1803 - val_accuracy: 0.9375\n",
            "Epoch 69/120\n",
            "12/12 [==============================] - 9s 711ms/step - loss: 0.0326 - accuracy: 1.0000 - val_loss: 0.1682 - val_accuracy: 0.9583\n",
            "Epoch 70/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.0319 - accuracy: 1.0000 - val_loss: 0.1690 - val_accuracy: 0.9375\n",
            "Epoch 71/120\n",
            "12/12 [==============================] - 8s 711ms/step - loss: 0.0297 - accuracy: 1.0000 - val_loss: 0.1742 - val_accuracy: 0.9375\n",
            "Epoch 72/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0277 - accuracy: 1.0000 - val_loss: 0.1734 - val_accuracy: 0.9375\n",
            "Epoch 73/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0269 - accuracy: 1.0000 - val_loss: 0.1474 - val_accuracy: 0.9375\n",
            "Epoch 74/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0263 - accuracy: 1.0000 - val_loss: 0.1932 - val_accuracy: 0.9375\n",
            "Epoch 75/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0244 - accuracy: 1.0000 - val_loss: 0.2116 - val_accuracy: 0.9375\n",
            "Epoch 76/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0233 - accuracy: 1.0000 - val_loss: 0.1612 - val_accuracy: 0.9375\n",
            "Epoch 77/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0226 - accuracy: 1.0000 - val_loss: 0.1416 - val_accuracy: 0.9375\n",
            "Epoch 78/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0222 - accuracy: 1.0000 - val_loss: 0.1350 - val_accuracy: 0.9583\n",
            "Epoch 79/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.0218 - accuracy: 1.0000 - val_loss: 0.1403 - val_accuracy: 0.9583\n",
            "Epoch 80/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0218 - accuracy: 1.0000 - val_loss: 0.1447 - val_accuracy: 0.9375\n",
            "Epoch 81/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0204 - accuracy: 1.0000 - val_loss: 0.1486 - val_accuracy: 0.9583\n",
            "Epoch 82/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0193 - accuracy: 1.0000 - val_loss: 0.1655 - val_accuracy: 0.9583\n",
            "Epoch 83/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0184 - accuracy: 1.0000 - val_loss: 0.1495 - val_accuracy: 0.9583\n",
            "Epoch 84/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0181 - accuracy: 1.0000 - val_loss: 0.1495 - val_accuracy: 0.9583\n",
            "Epoch 85/120\n",
            "12/12 [==============================] - 8s 710ms/step - loss: 0.0172 - accuracy: 1.0000 - val_loss: 0.1356 - val_accuracy: 0.9375\n",
            "Epoch 86/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0168 - accuracy: 1.0000 - val_loss: 0.1489 - val_accuracy: 0.9583\n",
            "Epoch 87/120\n",
            "12/12 [==============================] - 8s 711ms/step - loss: 0.0162 - accuracy: 1.0000 - val_loss: 0.1435 - val_accuracy: 0.9792\n",
            "Epoch 88/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0153 - accuracy: 1.0000 - val_loss: 0.1150 - val_accuracy: 0.9583\n",
            "Epoch 89/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.0144 - accuracy: 1.0000 - val_loss: 0.1091 - val_accuracy: 0.9792\n",
            "Epoch 90/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.0139 - accuracy: 1.0000 - val_loss: 0.1131 - val_accuracy: 0.9792\n",
            "Epoch 91/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0133 - accuracy: 1.0000 - val_loss: 0.1106 - val_accuracy: 0.9792\n",
            "Epoch 92/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0129 - accuracy: 1.0000 - val_loss: 0.1109 - val_accuracy: 0.9792\n",
            "Epoch 93/120\n",
            "12/12 [==============================] - 9s 712ms/step - loss: 0.0129 - accuracy: 1.0000 - val_loss: 0.1135 - val_accuracy: 0.9583\n",
            "Epoch 94/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.0122 - accuracy: 1.0000 - val_loss: 0.1345 - val_accuracy: 0.9792\n",
            "Epoch 95/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0118 - accuracy: 1.0000 - val_loss: 0.1181 - val_accuracy: 0.9792\n",
            "Epoch 96/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0112 - accuracy: 1.0000 - val_loss: 0.1181 - val_accuracy: 0.9792\n",
            "Epoch 97/120\n",
            "12/12 [==============================] - 8s 710ms/step - loss: 0.0109 - accuracy: 1.0000 - val_loss: 0.1281 - val_accuracy: 0.9792\n",
            "Epoch 98/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0104 - accuracy: 1.0000 - val_loss: 0.1164 - val_accuracy: 0.9792\n",
            "Epoch 99/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0109 - accuracy: 1.0000 - val_loss: 0.1593 - val_accuracy: 0.9792\n",
            "Epoch 100/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0390 - accuracy: 0.9974 - val_loss: 0.2416 - val_accuracy: 0.9375\n",
            "Epoch 101/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0144 - accuracy: 1.0000 - val_loss: 0.0712 - val_accuracy: 0.9792\n",
            "Epoch 102/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0261 - accuracy: 0.9948 - val_loss: 0.1453 - val_accuracy: 0.9583\n",
            "Epoch 103/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0219 - accuracy: 0.9948 - val_loss: 0.1674 - val_accuracy: 0.9375\n",
            "Epoch 104/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0208 - accuracy: 0.9974 - val_loss: 0.1542 - val_accuracy: 0.9375\n",
            "Epoch 105/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0517 - accuracy: 0.9896 - val_loss: 0.2224 - val_accuracy: 0.9375\n",
            "Epoch 106/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.2843 - accuracy: 0.9297 - val_loss: 0.4335 - val_accuracy: 0.8958\n",
            "Epoch 107/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.3310 - accuracy: 0.8984 - val_loss: 0.4080 - val_accuracy: 0.8333\n",
            "Epoch 108/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.2126 - accuracy: 0.9401 - val_loss: 0.2548 - val_accuracy: 0.8958\n",
            "Epoch 109/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.1524 - accuracy: 0.9609 - val_loss: 0.3534 - val_accuracy: 0.8333\n",
            "Epoch 110/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0889 - accuracy: 0.9870 - val_loss: 0.2393 - val_accuracy: 0.9375\n",
            "Epoch 111/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.0658 - accuracy: 0.9896 - val_loss: 0.2299 - val_accuracy: 0.8958\n",
            "Epoch 112/120\n",
            "12/12 [==============================] - 9s 715ms/step - loss: 0.0767 - accuracy: 0.9844 - val_loss: 0.1895 - val_accuracy: 0.9167\n",
            "Epoch 113/120\n",
            "12/12 [==============================] - 8s 706ms/step - loss: 0.1420 - accuracy: 0.9714 - val_loss: 0.2288 - val_accuracy: 0.8958\n",
            "Epoch 114/120\n",
            "12/12 [==============================] - 8s 710ms/step - loss: 0.1209 - accuracy: 0.9714 - val_loss: 0.2296 - val_accuracy: 0.9167\n",
            "Epoch 115/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.0642 - accuracy: 0.9922 - val_loss: 0.1061 - val_accuracy: 0.9583\n",
            "Epoch 116/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0478 - accuracy: 1.0000 - val_loss: 0.1648 - val_accuracy: 0.9583\n",
            "Epoch 117/120\n",
            "12/12 [==============================] - 8s 710ms/step - loss: 0.0554 - accuracy: 0.9948 - val_loss: 0.1037 - val_accuracy: 0.9792\n",
            "Epoch 118/120\n",
            "12/12 [==============================] - 8s 707ms/step - loss: 0.0353 - accuracy: 1.0000 - val_loss: 0.1088 - val_accuracy: 0.9792\n",
            "Epoch 119/120\n",
            "12/12 [==============================] - 8s 708ms/step - loss: 0.0340 - accuracy: 1.0000 - val_loss: 0.0871 - val_accuracy: 0.9583\n",
            "Epoch 120/120\n",
            "12/12 [==============================] - 8s 709ms/step - loss: 0.0297 - accuracy: 1.0000 - val_loss: 0.0966 - val_accuracy: 0.9583\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BYbXLZEvje5n",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ed58af80-6aa8-495b-8288-cbad3b0cb8e0"
      },
      "source": [
        "start=time.time()\n",
        "score=model.evaluate([video_features_test,audio_features_test],labels_test)\n",
        "end=time.time()\n",
        "test_time=end-start\n",
        "print(\"Test loss of the model is - \" , score[0])\n",
        "print(\"Test accuracy of the model is - \" , score[1]*100 , \"%\")"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "2/2 [==============================] - 0s 134ms/step - loss: 0.1533 - accuracy: 0.9792\n",
            "Test loss of the model is -  0.15330761671066284\n",
            "Test accuracy of the model is -  97.91666865348816 %\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sS4hGoWVbQqH",
        "outputId": "e69f3e11-5391-43b6-d104-fbb206239cde"
      },
      "source": [
        "print(train_time_avg)\n",
        "print(test_time)"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "8.703442581494649\n",
            "0.5022244453430176\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XoT9pkjhFouF"
      },
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "predictions = model.predict([video_features_test,audio_features_test])\n",
        "predictions = predictions.argmax(axis=1)\n",
        "predictions = predictions.astype(int).flatten()\n",
        "predictions = (lb.inverse_transform((predictions)))\n",
        "actual = labels_test.argmax(axis=1)\n",
        "actual = actual.astype(int).flatten()\n",
        "actual = (lb.inverse_transform((actual)))\n",
        "cm = confusion_matrix(actual, predictions,labels=['a', 'd', 'f', 'h', 'n', 'sa', 'su'],normalize='true')\n",
        "conf_matrix= pd.DataFrame(cm,index=None,columns=None)\n",
        "conf_matrix.to_csv(r'confusion_matrix.csv', index = None, header=False)"
      ],
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "v1Q0Kl5kKmKU",
        "outputId": "7d374f4f-694f-4ddb-fb45-e87a58afdc8c"
      },
      "source": [
        "from sklearn.metrics import classification_report\n",
        "print(classification_report(actual, predictions, target_names = ['a', 'd', 'f', 'h', 'n', 'sa', 'su']))"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           a       1.00      1.00      1.00         6\n",
            "           d       1.00      1.00      1.00         3\n",
            "           f       1.00      1.00      1.00         4\n",
            "           h       0.86      1.00      0.92         6\n",
            "           n       1.00      1.00      1.00        13\n",
            "          sa       1.00      1.00      1.00         7\n",
            "          su       1.00      0.89      0.94         9\n",
            "\n",
            "    accuracy                           0.98        48\n",
            "   macro avg       0.98      0.98      0.98        48\n",
            "weighted avg       0.98      0.98      0.98        48\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FpJgBbGQMzJc",
        "outputId": "0b9310ee-50ca-40c0-832c-e613a0e64052"
      },
      "source": [
        "from imblearn.metrics import specificity_score\n",
        "s=specificity_score(actual,predictions,average='macro')\n",
        "print(s)"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.9965986394557823\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "itwLx4UmM4Ch",
        "outputId": "11283b0e-2609-4e35-8008-69d2c42fddc4"
      },
      "source": [
        "from sklearn.metrics import roc_auc_score\n",
        "y_score=model.predict([video_features_test,audio_features_test])\n",
        "auc=roc_auc_score(labels_test, y_score, average='macro',multi_class='ovo')\n",
        "print(auc)"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.998459212744927\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PW6PC3lgM9BH",
        "outputId": "003b039c-9e24-4911-a9de-b553cca54345"
      },
      "source": [
        "from sklearn.metrics import precision_score\n",
        "from sklearn.metrics import recall_score\n",
        "from sklearn.metrics import f1_score\n",
        "p=precision_score(actual,predictions,average='macro')\n",
        "r=recall_score(actual,predictions,average='macro')\n",
        "f=f1_score(actual,predictions,average='macro')\n",
        "print(p,r,f)"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.979591836734694 0.9841269841269842 0.9806076276664513\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YyrWTl4qOYYG"
      },
      "source": [
        "from numpy import interp\n",
        "from sklearn.metrics import roc_curve\n",
        "fpr = dict()\n",
        "tpr = dict()\n",
        "classes=7\n",
        "y_score=model.predict([video_features_test,audio_features_test])\n",
        "for i in range(classes):\n",
        "  fpr[i], tpr[i], _ = roc_curve(labels_test[:, i], y_score[:, i])\n",
        "\n",
        "all_fpr = np.unique(np.concatenate([fpr[i] for i in range(classes)]))\n",
        "mean_tpr = np.zeros_like(all_fpr)\n",
        "for i in range(classes):\n",
        "    mean_tpr += interp(all_fpr, fpr[i], tpr[i])\n",
        "mean_tpr /= classes\n",
        "\n",
        "fpr['macro'] = all_fpr\n",
        "tpr['macro'] = mean_tpr"
      ],
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jRZE4XLjObBu"
      },
      "source": [
        "l=[]\n",
        "for i in range(120):\n",
        "  l.append(i+1)"
      ],
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "O_QR33_8OkUR",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1bf9cd7b-0594-4493-842b-0e5ebe8f2803"
      },
      "source": [
        "from pandas import DataFrame\n",
        "C = {'epoch no':l,\n",
        "    'acc train':history.history['accuracy'],\n",
        "    'acc val':history.history['val_accuracy'],\n",
        "    'loss train':history.history['loss'],\n",
        "    'loss val':history.history['val_loss']}\n",
        "df = DataFrame(C, columns= ['epoch no','acc train','acc val','loss train','loss val'])\n",
        "export_csv = df.to_csv (r'result1.csv', index = None, header=True) \n",
        "print (df)"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "     epoch no  acc train   acc val  loss train  loss val\n",
            "0           1   0.315104  0.333333    2.647303  2.353148\n",
            "1           2   0.359375  0.416667    2.251528  2.131384\n",
            "2           3   0.369792  0.416667    1.993414  1.940153\n",
            "3           4   0.393229  0.437500    1.842656  1.728528\n",
            "4           5   0.427083  0.458333    1.711660  1.624866\n",
            "..        ...        ...       ...         ...       ...\n",
            "115       116   1.000000  0.958333    0.047769  0.164767\n",
            "116       117   0.994792  0.979167    0.055410  0.103714\n",
            "117       118   1.000000  0.979167    0.035255  0.108767\n",
            "118       119   1.000000  0.958333    0.034038  0.087095\n",
            "119       120   1.000000  0.958333    0.029697  0.096566\n",
            "\n",
            "[120 rows x 5 columns]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nH2KPkGpOt8r",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ff9a479d-db85-45f5-90b1-f694a8885b59"
      },
      "source": [
        "C = {\n",
        "     'train acc':[str(history.history['accuracy'][119])],\n",
        "     'test acc':[str(score[1])],\n",
        "     'precision(macro)':[str(p)],\n",
        "     'recall(macro)':[str(r)],\n",
        "     'f1(macro)':[str(f)],\n",
        "     'specificity(macro)':[str(s)],\n",
        "     'AUC(macro)':[str(auc)],\n",
        "     'average train time(s)':[str(train_time_avg)],\n",
        "     'test time(s)':[str(test_time)],}\n",
        "df = DataFrame(C, columns= ['train acc','test acc','precision(macro)','recall(macro)','f1(macro)','specificity(macro)','AUC(macro)','average train time(s)','test time(s)'])\n",
        "export_csv = df.to_csv (r'result2.csv', index = None, header=True) \n",
        "print (df)"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "  train acc            test acc  ... average train time(s)        test time(s)\n",
            "0       1.0  0.9791666865348816  ...     8.703442581494649  0.5022244453430176\n",
            "\n",
            "[1 rows x 9 columns]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z4s-NTT6QxqT",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d4bdeb39-7960-47e3-8f31-7a3146e31996"
      },
      "source": [
        "C = {\n",
        "     'TPR(macro)': tpr['macro'],\n",
        "     'FPR(macro)': fpr['macro']}\n",
        "df = DataFrame(C, columns= ['TPR(macro)','FPR(macro)'])\n",
        "export_csv = df.to_csv (r'result3.csv', index = None, header=True) \n",
        "print (df)"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   TPR(macro)  FPR(macro)\n",
            "0    0.936508    0.000000\n",
            "1    0.984127    0.023810\n",
            "2    1.000000    0.025641\n",
            "3    1.000000    1.000000\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}