emotion_detection.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "emotion_detection.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyMQZu4w1+SuB7w1BVk+SvaX",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Varun221/Real_time_emotion_detection/blob/master/emotion_detection.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kDabKvWoTeWW",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# My numpy arrays are stored in google drive. \n",
        " \n",
        "from pydrive.auth import GoogleAuth\n",
        "from pydrive.drive import GoogleDrive\n",
        "from google.colab import auth\n",
        "from oauth2client.client import GoogleCredentials\n",
        " \n",
        "# 1. Authenticate and create the PyDrive client.\n",
        "auth.authenticate_user()\n",
        "gauth = GoogleAuth()\n",
        "gauth.credentials = GoogleCredentials.get_application_default()\n",
        "drive = GoogleDrive(gauth)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5xDl1VpnTliI",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# 2. Get the file\n",
        "def get_file_from_drive(folder_id, file_name):\n",
        "  file_list = drive.ListFile({'q': \"'\" + folder_id + \"' in parents and trashed=false\"}).GetList()\n",
        "  for file in file_list:\n",
        "    if file['title'] == file_name:\n",
        "      return file['id']\n",
        " \n",
        " \n",
        "drive_folder_id = '1oj3k3kGIpqaYfCZpJ2aqV5f3yWzQFeVC'\n",
        " \n",
        "def upload_data_system():\n",
        "  downloaded = drive.CreateFile({'id': get_file_from_drive(drive_folder_id, 'X_train.npy')})\n",
        "  downloaded.GetContentFile('X_train.npy') \n",
        " \n",
        "  downloaded = drive.CreateFile({'id': get_file_from_drive(drive_folder_id, 'Y_train.npy')})\n",
        "  downloaded.GetContentFile('Y_train.npy')\n",
        " \n",
        "  downloaded = drive.CreateFile({'id': get_file_from_drive(drive_folder_id, 'X_test.npy')})\n",
        "  downloaded.GetContentFile('X_test.npy')\n",
        " \n",
        "  downloaded = drive.CreateFile({'id': get_file_from_drive(drive_folder_id, 'Y_test.npy')})\n",
        "  downloaded.GetContentFile('Y_test.npy')\n",
        " \n",
        " \n",
        " \n",
        "upload_data_system()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PBU2-RtLT898",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# importing the libraries\n",
        "%tensorflow_version 2.x\n",
        "import numpy as np\n",
        "import pickle\n",
        "import pandas as pd\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "import cv2\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IOh2XNCMUMOa",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# mapping emotion to index\n",
        "emotion_to_index = {'angry':0,\n",
        "                   'disgusted':1,\n",
        "                   'fearful':2,\n",
        "                   'happy':3,\n",
        "                   'neutral':4,\n",
        "                   'sad':5,\n",
        "                   'surprised':6}\n",
        "index_to_emotion = {0: 'angry',\n",
        "                    1: 'disgusted',\n",
        "                    2: 'fearful',\n",
        "                    3: 'happy',\n",
        "                    4: 'neutral',\n",
        "                    5: 'sad',\n",
        "                    6: 'surprised'}\n",
        " \n",
        "emotions = list(emotion_to_index.keys())"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SBPsSmHNUUoC",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "3fcf4b1f-e4da-438c-aab0-1f856ed0ca64"
      },
      "source": [
        "# Lets load up the files\n",
        "X_train = np.load('X_train.npy')\n",
        "X_test = np.load('X_test.npy')\n",
        "Y_train = np.load('Y_train.npy')\n",
        "Y_test = np.load('Y_test.npy')\n",
        "X_train.shape, Y_train.shape, X_test.shape, Y_test.shape"
      ],
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((28709, 48, 48, 3), (28709, 1), (7178, 48, 48, 3), (7178, 1))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PIMU5zdaUxS_",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 284
        },
        "outputId": "f90e412d-05b6-4f2d-dd7d-6dc2cf907565"
      },
      "source": [
        "# visualizing the files based on index\n",
        "index = 1067\n",
        "img = X_train[index,:,:,:]\n",
        "print(f\"This is {index_to_emotion[Y_train[index, 0]]}\")\n",
        "img = img.astype(np.uint8)\n",
        "plt.imshow(img, cmap='Greys')\n",
        "plt.show()"
      ],
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "This is happy\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nVgazLFVVYLE",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 302
        },
        "outputId": "6960f2c1-d679-463d-98f5-710de3906db7"
      },
      "source": [
        "# number of images pertaining to each class\n",
        "arr = np.zeros((7,))\n",
        "for i in range(X_train.shape[0]):\n",
        "    val = np.uint8(Y_train[i])\n",
        "    arr[val] = arr[val] + 1\n",
        "#for i in range(7):\n",
        "#  print(f\"{index_to_emotion[i]} has {arr[i,0]}\")\n",
        "arr2 = np.reshape(arr, (-1))\n",
        "plt.bar([0,1,2,3,4,5,6], arr2, color='green')\n",
        "plt.xlabel(\"Emotion\", fontsize = 15)\n",
        "plt.ylabel(\"Number of images\", fontsize = 15)\n",
        "plt.title(\"Number of image per emotion in training set\", fontsize=15)\n",
        " \n",
        "plt.xticks([0,1,2,3,4,5,6], emotions)\n",
        " \n",
        "plt.show()"
      ],
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IAf1er60Y-_z",
        "colab_type": "text"
      },
      "source": [
        "As we can see, the number of images for the emotion disgust is quite low and the distribution for images is not quite same. Hence we will be using f1 score of model as an evaluation metric."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L_G_MN32V36X",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "3a2db8ad-f37f-4044-c616-81280162cbf8"
      },
      "source": [
        "# preproceesing the images\n",
        "# scaling the X arrays\n",
        "X_train_proc = X_train/255.\n",
        "X_test_proc = X_test/255.\n",
        "# one hot encoding the Y arrays\n",
        "Y_train_proc = np.squeeze(np.array(tf.one_hot(Y_train,7, axis=1)))\n",
        "Y_test_proc = np.squeeze(np.array(tf.one_hot(Y_test,7, axis=1)))\n",
        "Y_train_proc.shape, Y_test_proc.shape"
      ],
      "execution_count": 73,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((28709, 7), (7178, 7))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 73
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eK1xpwXYZv7t",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from tensorflow.keras import Sequential\n",
        "from tensorflow.keras import Model\n",
        "from tensorflow.keras.layers import Conv2D, Dense, Flatten, BatchNormalization, Dropout, MaxPooling2D, Input"
      ],
      "execution_count": 74,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9DsqtSbgfx1K",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "input = Input(shape=(48,48,3))\n",
        "x = Conv2D(16, (5,5), activation='relu')(input)\n",
        "x = BatchNormalization()(x)\n",
        "x = MaxPooling2D((3,3), strides=2)(x)\n",
        "x = Dropout(0.25)(x)\n",
        "x = Conv2D(32, (5,5), activation='relu')(x)\n",
        "x = MaxPooling2D((3,3), strides=2)(x)\n",
        "x = Dropout(0.2)(x)\n",
        "x = Conv2D(64, (4,4), activation='relu')(x)\n",
        "x = Flatten()(x)\n",
        "x = Dense(40, activation='relu')(x)\n",
        "output = Dense(7, activation='softmax')(x)\n",
        "model = Model(inputs=input, outputs=output)"
      ],
      "execution_count": 76,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cDWxsX5QZ6Dw",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 561
        },
        "outputId": "20ba91d3-09fd-4300-9b8e-c295ee822bfa"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"model_8\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "input_10 (InputLayer)        [(None, 48, 48, 3)]       0         \n",
            "_________________________________________________________________\n",
            "conv2d_25 (Conv2D)           (None, 44, 44, 16)        1216      \n",
            "_________________________________________________________________\n",
            "batch_normalization_9 (Batch (None, 44, 44, 16)        64        \n",
            "_________________________________________________________________\n",
            "max_pooling2d_17 (MaxPooling (None, 21, 21, 16)        0         \n",
            "_________________________________________________________________\n",
            "dropout_23 (Dropout)         (None, 21, 21, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_26 (Conv2D)           (None, 17, 17, 32)        12832     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_18 (MaxPooling (None, 8, 8, 32)          0         \n",
            "_________________________________________________________________\n",
            "dropout_24 (Dropout)         (None, 8, 8, 32)          0         \n",
            "_________________________________________________________________\n",
            "conv2d_27 (Conv2D)           (None, 5, 5, 64)          32832     \n",
            "_________________________________________________________________\n",
            "flatten_8 (Flatten)          (None, 1600)              0         \n",
            "_________________________________________________________________\n",
            "dense_16 (Dense)             (None, 40)                64040     \n",
            "_________________________________________________________________\n",
            "dense_17 (Dense)             (None, 7)                 287       \n",
            "=================================================================\n",
            "Total params: 111,271\n",
            "Trainable params: 111,239\n",
            "Non-trainable params: 32\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hsA1rTJJdiAA",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# compiling the model with adam optimizer\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "opt = Adam(lr=0.0001, decay=0.0001 / 40)\n",
        " \n",
        "model.compile(optimizer=opt, loss = 'categorical_crossentropy', metrics = [tf.keras.metrics.Precision(),\n",
        "                                                                           tf.keras.metrics.Recall(),\n",
        "                                                                           'accuracy'])"
      ],
      "execution_count": 78,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-UH1V7l-jp9n",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "a79c83e7-8453-4ab4-801b-271f06df76c6"
      },
      "source": [
        "history = model.fit(X_train_proc, Y_train_proc, epochs=30, verbose=1, validation_data=(X_test_proc, Y_test_proc))"
      ],
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/30\n",
            "898/898 [==============================] - 12s 13ms/step - loss: 1.7322 - precision_7: 0.6238 - recall_7: 0.0328 - accuracy: 0.3036 - val_loss: 1.5999 - val_precision_7: 0.7661 - val_recall_7: 0.0548 - val_accuracy: 0.3948\n",
            "Epoch 2/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.5687 - precision_7: 0.6856 - recall_7: 0.1272 - accuracy: 0.3924 - val_loss: 1.4985 - val_precision_7: 0.7816 - val_recall_7: 0.1371 - val_accuracy: 0.4369\n",
            "Epoch 3/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.4987 - precision_7: 0.7033 - recall_7: 0.1704 - accuracy: 0.4206 - val_loss: 1.4474 - val_precision_7: 0.7870 - val_recall_7: 0.1631 - val_accuracy: 0.4465\n",
            "Epoch 4/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.4367 - precision_7: 0.7201 - recall_7: 0.2048 - accuracy: 0.4480 - val_loss: 1.4138 - val_precision_7: 0.7894 - val_recall_7: 0.1921 - val_accuracy: 0.4645\n",
            "Epoch 5/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.3883 - precision_7: 0.7261 - recall_7: 0.2298 - accuracy: 0.4650 - val_loss: 1.3562 - val_precision_7: 0.7955 - val_recall_7: 0.2200 - val_accuracy: 0.4905\n",
            "Epoch 6/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.3537 - precision_7: 0.7288 - recall_7: 0.2524 - accuracy: 0.4835 - val_loss: 1.3821 - val_precision_7: 0.8049 - val_recall_7: 0.1885 - val_accuracy: 0.4788\n",
            "Epoch 7/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.3199 - precision_7: 0.7311 - recall_7: 0.2716 - accuracy: 0.4987 - val_loss: 1.3440 - val_precision_7: 0.8005 - val_recall_7: 0.2113 - val_accuracy: 0.5014\n",
            "Epoch 8/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.2940 - precision_7: 0.7320 - recall_7: 0.2869 - accuracy: 0.5080 - val_loss: 1.3374 - val_precision_7: 0.8015 - val_recall_7: 0.2104 - val_accuracy: 0.4967\n",
            "Epoch 9/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.2703 - precision_7: 0.7355 - recall_7: 0.3006 - accuracy: 0.5173 - val_loss: 1.2764 - val_precision_7: 0.8022 - val_recall_7: 0.2581 - val_accuracy: 0.5206\n",
            "Epoch 10/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.2458 - precision_7: 0.7409 - recall_7: 0.3143 - accuracy: 0.5245 - val_loss: 1.2797 - val_precision_7: 0.7023 - val_recall_7: 0.3217 - val_accuracy: 0.5106\n",
            "Epoch 11/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.2293 - precision_7: 0.7428 - recall_7: 0.3255 - accuracy: 0.5329 - val_loss: 1.2365 - val_precision_7: 0.7471 - val_recall_7: 0.3271 - val_accuracy: 0.5309\n",
            "Epoch 12/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.2139 - precision_7: 0.7432 - recall_7: 0.3342 - accuracy: 0.5402 - val_loss: 1.2533 - val_precision_7: 0.7865 - val_recall_7: 0.2895 - val_accuracy: 0.5205\n",
            "Epoch 13/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.1977 - precision_7: 0.7446 - recall_7: 0.3427 - accuracy: 0.5464 - val_loss: 1.2257 - val_precision_7: 0.7541 - val_recall_7: 0.3337 - val_accuracy: 0.5346\n",
            "Epoch 14/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.1784 - precision_7: 0.7472 - recall_7: 0.3536 - accuracy: 0.5514 - val_loss: 1.2067 - val_precision_7: 0.7776 - val_recall_7: 0.3321 - val_accuracy: 0.5407\n",
            "Epoch 15/30\n",
            "898/898 [==============================] - 12s 13ms/step - loss: 1.1656 - precision_7: 0.7494 - recall_7: 0.3639 - accuracy: 0.5609 - val_loss: 1.1953 - val_precision_7: 0.7716 - val_recall_7: 0.3346 - val_accuracy: 0.5469\n",
            "Epoch 16/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.1446 - precision_7: 0.7512 - recall_7: 0.3785 - accuracy: 0.5709 - val_loss: 1.1914 - val_precision_7: 0.7866 - val_recall_7: 0.3323 - val_accuracy: 0.5443\n",
            "Epoch 17/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.1378 - precision_7: 0.7462 - recall_7: 0.3831 - accuracy: 0.5691 - val_loss: 1.1906 - val_precision_7: 0.7811 - val_recall_7: 0.3356 - val_accuracy: 0.5485\n",
            "Epoch 18/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.1224 - precision_7: 0.7569 - recall_7: 0.3919 - accuracy: 0.5810 - val_loss: 1.1914 - val_precision_7: 0.7466 - val_recall_7: 0.3608 - val_accuracy: 0.5408\n",
            "Epoch 19/30\n",
            "898/898 [==============================] - 12s 13ms/step - loss: 1.1098 - precision_7: 0.7546 - recall_7: 0.4007 - accuracy: 0.5822 - val_loss: 1.1872 - val_precision_7: 0.7542 - val_recall_7: 0.3590 - val_accuracy: 0.5525\n",
            "Epoch 20/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.0928 - precision_7: 0.7585 - recall_7: 0.4120 - accuracy: 0.5899 - val_loss: 1.1994 - val_precision_7: 0.7602 - val_recall_7: 0.3502 - val_accuracy: 0.5472\n",
            "Epoch 21/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.0864 - precision_7: 0.7562 - recall_7: 0.4151 - accuracy: 0.5919 - val_loss: 1.1561 - val_precision_7: 0.7601 - val_recall_7: 0.3743 - val_accuracy: 0.5655\n",
            "Epoch 22/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 1.0728 - precision_7: 0.7586 - recall_7: 0.4229 - accuracy: 0.5973 - val_loss: 1.1604 - val_precision_7: 0.7505 - val_recall_7: 0.3788 - val_accuracy: 0.5568\n",
            "Epoch 23/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.0638 - precision_7: 0.7596 - recall_7: 0.4284 - accuracy: 0.5995 - val_loss: 1.1545 - val_precision_7: 0.7408 - val_recall_7: 0.3862 - val_accuracy: 0.5599\n",
            "Epoch 24/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.0528 - precision_7: 0.7624 - recall_7: 0.4372 - accuracy: 0.6056 - val_loss: 1.1511 - val_precision_7: 0.7642 - val_recall_7: 0.3778 - val_accuracy: 0.5671\n",
            "Epoch 25/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.0377 - precision_7: 0.7624 - recall_7: 0.4480 - accuracy: 0.6120 - val_loss: 1.1654 - val_precision_7: 0.7178 - val_recall_7: 0.4040 - val_accuracy: 0.5548\n",
            "Epoch 26/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.0285 - precision_7: 0.7619 - recall_7: 0.4519 - accuracy: 0.6135 - val_loss: 1.1593 - val_precision_7: 0.7524 - val_recall_7: 0.3827 - val_accuracy: 0.5600\n",
            "Epoch 27/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.0165 - precision_7: 0.7689 - recall_7: 0.4652 - accuracy: 0.6214 - val_loss: 1.1527 - val_precision_7: 0.7333 - val_recall_7: 0.4057 - val_accuracy: 0.5662\n",
            "Epoch 28/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 1.0095 - precision_7: 0.7695 - recall_7: 0.4675 - accuracy: 0.6261 - val_loss: 1.1643 - val_precision_7: 0.7217 - val_recall_7: 0.4154 - val_accuracy: 0.5620\n",
            "Epoch 29/30\n",
            "898/898 [==============================] - 11s 12ms/step - loss: 0.9998 - precision_7: 0.7685 - recall_7: 0.4733 - accuracy: 0.6251 - val_loss: 1.1751 - val_precision_7: 0.7318 - val_recall_7: 0.4014 - val_accuracy: 0.5577\n",
            "Epoch 30/30\n",
            "898/898 [==============================] - 11s 13ms/step - loss: 0.9866 - precision_7: 0.7722 - recall_7: 0.4825 - accuracy: 0.6326 - val_loss: 1.1496 - val_precision_7: 0.7527 - val_recall_7: 0.3944 - val_accuracy: 0.5695\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TGrG87j-j8P3",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "737201c7-3dc5-407c-ff1f-5ecb0d1cf398"
      },
      "source": [
        "# PLOTTING TRAINING RESULTS\n",
        "# plotting accuracy, loss, recall, precision and f1 score of training and validation sets\n",
        " \n",
        "num = 7\n",
        "prec = f\"precision_{num}\"\n",
        "rec = f\"recall_{num}\"\n",
        "valprec = f\"val_precision_{num}\"\n",
        "valrec = f\"val_recall_{num}\"\n",
        " \n",
        " \n",
        "plt.figure(figsize=(10,20))\n",
        " \n",
        "plt.subplot(511)\n",
        "plt.plot(history.history['accuracy'], '--', label='acc')\n",
        "plt.plot(history.history['val_accuracy'], '--',label='val_acc')\n",
        "plt.title('model accuracy',fontsize=15, color='green')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(loc='upper left')\n",
        "#axes=plt.gca()\n",
        "#axes.set_ylim([0.965, 1])\n",
        "plt.tight_layout(h_pad=3.0)\n",
        " \n",
        " \n",
        "plt.subplot(512)\n",
        "plt.plot(history.history['loss'], '--', label='loss')\n",
        "plt.plot(history.history['val_loss'], '--',label='val_loss')\n",
        "plt.title('model loss',fontsize=15, color='green')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(loc='upper left')\n",
        "#axes=plt.gca()\n",
        " \n",
        "plt.subplot(513)\n",
        "plt.plot(history.history[prec], '--', label='precision')\n",
        "plt.plot(history.history[valprec], '--',label='val_precision')\n",
        "plt.title('model precision',fontsize=15, color='green')\n",
        "plt.ylabel('precision')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(loc='upper left')\n",
        "#axes=plt.gca()\n",
        " \n",
        "plt.subplot(514)\n",
        "plt.plot(history.history[rec], '--', label='recall')\n",
        "plt.plot(history.history[valrec], '--',label='val_recall')\n",
        "plt.title('model recall',fontsize=15, color='green')\n",
        "plt.ylabel('recall')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(loc='upper left')\n",
        "#axes=plt.gca()\n",
        " \n",
        "precision = history.history[prec]\n",
        "recall = history.history[rec]\n",
        "precval = history.history[valprec]\n",
        "recval = history.history[valrec]\n",
        "f1 = [(2*a*b)/(a+b) for a,b in zip(precision, recall)]\n",
        "f1_val = [(2*a*b)/(a+b+1e-7) for a,b in zip(precval, recval)]\n",
        " \n",
        "plt.subplot(515)\n",
        "plt.plot(f1, '--', label='f1_score')\n",
        "plt.plot(f1_val, '--',label='f1_score_val')\n",
        "plt.title('model f1 score',fontsize=15,color='green')\n",
        "plt.ylabel('f1 score')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(loc='upper left')\n",
        "#axes=plt.gca()\n",
        " \n",
        "plt.show()"
      ],
      "execution_count": 80,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x1440 with 5 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CCHvhQ1zpBpi",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 459
        },
        "outputId": "a7852c54-9984-4f80-b6b3-0b052031e0e1"
      },
      "source": [
        "# classification reports and confusion matrix\n",
        "from sklearn.metrics import classification_report\n",
        "print('\\nClassification Report\\n')\n",
        "preds = model.predict(X_test_proc)\n",
        "y_pred = np.argmax(preds, axis=1)\n",
        "print(classification_report(Y_test, y_pred, target_names=emotions))\n",
        "from sklearn.metrics import confusion_matrix\n",
        "confusion = confusion_matrix(Y_test, y_pred)\n",
        "print('Confusion Matrix\\n')\n",
        "print(confusion)"
      ],
      "execution_count": 81,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Classification Report\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "       angry       0.45      0.53      0.48       958\n",
            "   disgusted       0.56      0.28      0.37       111\n",
            "     fearful       0.48      0.28      0.35      1024\n",
            "       happy       0.73      0.82      0.77      1774\n",
            "     neutral       0.47      0.60      0.53      1233\n",
            "         sad       0.47      0.36      0.40      1247\n",
            "   surprised       0.72      0.75      0.73       831\n",
            "\n",
            "    accuracy                           0.57      7178\n",
            "   macro avg       0.55      0.52      0.52      7178\n",
            "weighted avg       0.56      0.57      0.56      7178\n",
            "\n",
            "Confusion Matrix\n",
            "\n",
            "[[ 503    9   70   90  158  102   26]\n",
            " [  45   31    9    9   10    5    2]\n",
            " [ 152    5  284   91  195  176  121]\n",
            " [  65    0   38 1461  120   62   28]\n",
            " [ 134    5   46  120  745  148   35]\n",
            " [ 199    3   87  168  315  444   31]\n",
            " [  26    2   53   62   52   16  620]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qnpbYgwE3NrM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 910
        },
        "outputId": "d5acf8b4-8d38-487f-f0a3-706a1e011bf7"
      },
      "source": [
        "# confusion matrix\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt     \n",
        " \n",
        "plt.figure(figsize=(15,15))\n",
        "ax= plt.subplot()\n",
        "sns.heatmap(confusion, annot=True, ax = ax, fmt='g'); #annot=True to annotate cells\n",
        "ax.xaxis.set_tick_params(labeltop='on')\n",
        "# labels, title and ticks\n",
        "ax.set_xlabel('Predicted labels');\n",
        "ax.set_ylabel('True labels'); \n",
        "ax.set_title('Confusion Matrix'); \n",
        "rev = []\n",
        "for i in range(6,-1,-1):\n",
        "  rev.append(emotions[i])\n",
        "ax.xaxis.set_ticklabels(emotions); \n",
        " \n",
        "ax.yaxis.set_ticklabels(emotions);"
      ],
      "execution_count": 82,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x1080 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_Nc1FDpV7JMd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# saving the model\n",
        "model.save('my_model.h5')"
      ],
      "execution_count": 83,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FiiJcavJHRiP",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}