FER_model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "dHlV0B0XLdXz",
    "outputId": "bc8f1167-afbb-4e70-c9d6-5e8ec5e154ed"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mounted at /content/gdrive\n"
     ]
    }
   ],
   "source": [
    "from google.colab import drive\n",
    "drive.mount(\"/content/gdrive\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "NR6HdOwvVUsY",
    "outputId": "bafc15fd-780e-4c42-e72c-95690206f1bf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done\n"
     ]
    }
   ],
   "source": [
    "from zipfile import ZipFile\n",
    "file_name = \"/content/gdrive/MyDrive/Colab Notebooks/FER/archive.zip\"\n",
    "\n",
    "with ZipFile(file_name,'r') as zip:\n",
    "  zip.extractall()\n",
    "  print(\"done\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "H7YqEfUvhWE1"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import cv2\n",
    "import tensorflow\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Flatten, Dropout,Conv2D,MaxPooling2D\n",
    "from keras.preprocessing.image import ImageDataGenerator\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "KJZhoOeKVoD8",
    "outputId": "8d737ed5-22cb-4856-fa1d-e75387c4169e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 28821 images belonging to 7 classes.\n",
      "Found 7066 images belonging to 7 classes.\n"
     ]
    }
   ],
   "source": [
    "folder_path = \"/content/images/\"\n",
    "picture_size = 48\n",
    "batch_size = 64\n",
    "\n",
    "datagen_train = ImageDataGenerator(rescale=1./255);\n",
    "datagen_val = ImageDataGenerator(rescale=1./255);\n",
    "\n",
    "train_set = datagen_train.flow_from_directory(\n",
    "    folder_path + \"train\",\n",
    "    target_size = (picture_size, picture_size),\n",
    "    batch_size = batch_size,\n",
    "    color_mode = \"grayscale\",\n",
    "    class_mode = \"categorical\",\n",
    "    shuffle = True\n",
    ")\n",
    "\n",
    "validation_set = datagen_val.flow_from_directory(\n",
    "    folder_path + \"validation\",\n",
    "    target_size = (picture_size, picture_size),\n",
    "    batch_size = batch_size,\n",
    "    color_mode = \"grayscale\",\n",
    "    class_mode = \"categorical\",\n",
    "    shuffle = False\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "ktchi8yiljIE"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 46, 46, 32)        320       \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 44, 44, 64)        18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 22, 22, 64)        0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 22, 22, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 20, 20, 128)       73856     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 10, 10, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 8, 8, 128)         147584    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 4, 4, 128)         0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 4, 4, 128)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 2048)              0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1024)              2098176   \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 1024)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 7)                 7175      \n",
      "=================================================================\n",
      "Total params: 2,345,607\n",
      "Trainable params: 2,345,607\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "emotion_model = Sequential()\n",
    "emotion_model.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=(48,48,1)))\n",
    "emotion_model.add(Conv2D(64, kernel_size=(3, 3), activation='relu'))\n",
    "emotion_model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "emotion_model.add(Dropout(0.25))\n",
    "emotion_model.add(Conv2D(128, kernel_size=(3, 3), activation='relu'))\n",
    "emotion_model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "emotion_model.add(Conv2D(128, kernel_size=(3, 3), activation='relu'))\n",
    "emotion_model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "emotion_model.add(Dropout(0.25))\n",
    "emotion_model.add(Flatten())\n",
    "emotion_model.add(Dense(1024, activation='relu'))\n",
    "emotion_model.add(Dropout(0.5))\n",
    "emotion_model.add(Dense(7, activation='softmax'))\n",
    "emotion_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "uMzFg_5PmNtf",
    "outputId": "0681d7de-1e6e-4702-9847-2bf3e5927caa"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "450/450 [==============================] - 20s 40ms/step - loss: 1.7972 - accuracy: 0.2585 - val_loss: 1.6839 - val_accuracy: 0.3557\n",
      "Epoch 2/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.6284 - accuracy: 0.3671 - val_loss: 1.5540 - val_accuracy: 0.4178\n",
      "Epoch 3/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.5304 - accuracy: 0.4111 - val_loss: 1.4777 - val_accuracy: 0.4413\n",
      "Epoch 4/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.4565 - accuracy: 0.4439 - val_loss: 1.3994 - val_accuracy: 0.4672\n",
      "Epoch 5/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.3997 - accuracy: 0.4677 - val_loss: 1.3415 - val_accuracy: 0.4899\n",
      "Epoch 6/50\n",
      "450/450 [==============================] - 18s 39ms/step - loss: 1.3440 - accuracy: 0.4888 - val_loss: 1.3082 - val_accuracy: 0.5037\n",
      "Epoch 7/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.3030 - accuracy: 0.5064 - val_loss: 1.2659 - val_accuracy: 0.5185\n",
      "Epoch 8/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.2641 - accuracy: 0.5214 - val_loss: 1.2382 - val_accuracy: 0.5286\n",
      "Epoch 9/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.2262 - accuracy: 0.5357 - val_loss: 1.2161 - val_accuracy: 0.5420\n",
      "Epoch 10/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.1960 - accuracy: 0.5504 - val_loss: 1.1886 - val_accuracy: 0.5557\n",
      "Epoch 11/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.1664 - accuracy: 0.5616 - val_loss: 1.1788 - val_accuracy: 0.5608\n",
      "Epoch 12/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.1350 - accuracy: 0.5767 - val_loss: 1.1599 - val_accuracy: 0.5632\n",
      "Epoch 13/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.1123 - accuracy: 0.5834 - val_loss: 1.1665 - val_accuracy: 0.5624\n",
      "Epoch 14/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.0855 - accuracy: 0.5958 - val_loss: 1.1319 - val_accuracy: 0.5824\n",
      "Epoch 15/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.0564 - accuracy: 0.6051 - val_loss: 1.1228 - val_accuracy: 0.5801\n",
      "Epoch 16/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.0306 - accuracy: 0.6150 - val_loss: 1.1134 - val_accuracy: 0.5822\n",
      "Epoch 17/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 1.0039 - accuracy: 0.6277 - val_loss: 1.1046 - val_accuracy: 0.5916\n",
      "Epoch 18/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.9814 - accuracy: 0.6387 - val_loss: 1.1042 - val_accuracy: 0.5911\n",
      "Epoch 19/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.9618 - accuracy: 0.6445 - val_loss: 1.0863 - val_accuracy: 0.5956\n",
      "Epoch 20/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.9328 - accuracy: 0.6544 - val_loss: 1.0882 - val_accuracy: 0.6010\n",
      "Epoch 21/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.9075 - accuracy: 0.6686 - val_loss: 1.0846 - val_accuracy: 0.6016\n",
      "Epoch 22/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.8793 - accuracy: 0.6766 - val_loss: 1.0780 - val_accuracy: 0.6043\n",
      "Epoch 23/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.8610 - accuracy: 0.6855 - val_loss: 1.0901 - val_accuracy: 0.6087\n",
      "Epoch 24/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.8342 - accuracy: 0.6921 - val_loss: 1.0753 - val_accuracy: 0.6104\n",
      "Epoch 25/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.8136 - accuracy: 0.7015 - val_loss: 1.0667 - val_accuracy: 0.6128\n",
      "Epoch 26/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.7888 - accuracy: 0.7149 - val_loss: 1.0834 - val_accuracy: 0.6081\n",
      "Epoch 27/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.7600 - accuracy: 0.7213 - val_loss: 1.0751 - val_accuracy: 0.6158\n",
      "Epoch 28/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.7341 - accuracy: 0.7308 - val_loss: 1.0818 - val_accuracy: 0.6163\n",
      "Epoch 29/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.7103 - accuracy: 0.7424 - val_loss: 1.0869 - val_accuracy: 0.6128\n",
      "Epoch 30/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.6925 - accuracy: 0.7476 - val_loss: 1.0986 - val_accuracy: 0.6179\n",
      "Epoch 31/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.6669 - accuracy: 0.7551 - val_loss: 1.0841 - val_accuracy: 0.6153\n",
      "Epoch 32/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.6459 - accuracy: 0.7655 - val_loss: 1.1035 - val_accuracy: 0.6220\n",
      "Epoch 33/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.6175 - accuracy: 0.7745 - val_loss: 1.1038 - val_accuracy: 0.6203\n",
      "Epoch 34/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.6095 - accuracy: 0.7797 - val_loss: 1.0967 - val_accuracy: 0.6187\n",
      "Epoch 35/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.5793 - accuracy: 0.7906 - val_loss: 1.1085 - val_accuracy: 0.6173\n",
      "Epoch 36/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.5618 - accuracy: 0.7975 - val_loss: 1.1177 - val_accuracy: 0.6186\n",
      "Epoch 37/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.5409 - accuracy: 0.8070 - val_loss: 1.1420 - val_accuracy: 0.6158\n",
      "Epoch 38/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.5205 - accuracy: 0.8106 - val_loss: 1.1271 - val_accuracy: 0.6210\n",
      "Epoch 39/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.5065 - accuracy: 0.8179 - val_loss: 1.1556 - val_accuracy: 0.6193\n",
      "Epoch 40/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.4884 - accuracy: 0.8231 - val_loss: 1.1542 - val_accuracy: 0.6189\n",
      "Epoch 41/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.4692 - accuracy: 0.8303 - val_loss: 1.1703 - val_accuracy: 0.6200\n",
      "Epoch 42/50\n",
      "450/450 [==============================] - 18s 39ms/step - loss: 0.4514 - accuracy: 0.8381 - val_loss: 1.1727 - val_accuracy: 0.6226\n",
      "Epoch 43/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.4378 - accuracy: 0.8438 - val_loss: 1.1709 - val_accuracy: 0.6185\n",
      "Epoch 44/50\n",
      "450/450 [==============================] - 17s 38ms/step - loss: 0.4281 - accuracy: 0.8468 - val_loss: 1.1825 - val_accuracy: 0.6190\n",
      "Epoch 45/50\n",
      "450/450 [==============================] - 18s 39ms/step - loss: 0.4117 - accuracy: 0.8514 - val_loss: 1.1779 - val_accuracy: 0.6172\n",
      "Epoch 46/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.3962 - accuracy: 0.8583 - val_loss: 1.2068 - val_accuracy: 0.6203\n",
      "Epoch 47/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.3814 - accuracy: 0.8649 - val_loss: 1.2228 - val_accuracy: 0.6175\n",
      "Epoch 48/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.3750 - accuracy: 0.8655 - val_loss: 1.2252 - val_accuracy: 0.6149\n",
      "Epoch 49/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.3597 - accuracy: 0.8709 - val_loss: 1.2261 - val_accuracy: 0.6233\n",
      "Epoch 50/50\n",
      "450/450 [==============================] - 17s 39ms/step - loss: 0.3538 - accuracy: 0.8735 - val_loss: 1.2294 - val_accuracy: 0.6213\n"
     ]
    }
   ],
   "source": [
    "emotion_model.compile(loss='categorical_crossentropy',optimizer=tensorflow.keras.optimizers.Adam(learning_rate=0.0001, decay=1e-6),metrics=['accuracy'])\n",
    "emotion_model_info = emotion_model.fit(\n",
    "        train_set,\n",
    "        steps_per_epoch=28821 // 64,\n",
    "        epochs=50,\n",
    "        validation_data=validation_set,\n",
    "        validation_steps=7066 // 64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "XUK27A5fmcu-"
   },
   "outputs": [],
   "source": [
    "emotion_model.save('model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 599
    },
    "id": "f009BxJkq6BA",
    "outputId": "eb251b5a-1a38-47a5-a9c6-bec332ab8842"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#accuracy\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.style.use('dark_background')\n",
    "\n",
    "plt.figure(figsize=(20,10))\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.suptitle('Optimizer : Adam', fontsize=10)\n",
    "plt.ylabel('Loss', fontsize=16)\n",
    "plt.plot(emotion_model_info.history['loss'], label='Training Loss')\n",
    "plt.plot(emotion_model_info.history['val_loss'], label='Validation Loss')\n",
    "plt.legend(loc='upper right')\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.ylabel('Accuracy', fontsize=16)\n",
    "plt.plot(emotion_model_info.history['accuracy'], label='Training Accuracy')\n",
    "plt.plot(emotion_model_info.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.legend(loc='lower right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "JUkvUbk_rvK-"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "name": "FER.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}