FacialExpressionRecognition(dir).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tensorflow Version : 2.1.0\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import utils\n",
    "%matplotlib inline\n",
    "\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras.layers import Dense, Input, Dropout, Flatten, Conv2D\n",
    "from tensorflow.keras.layers import BatchNormalization, Activation, MaxPooling2D\n",
    "from tensorflow.keras.models import Model, Sequential\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint, ReduceLROnPlateau\n",
    "from tensorflow.keras.utils import plot_model\n",
    "from tensorflow.keras.models import load_model, model_from_json\n",
    "from IPython.display import SVG, Image\n",
    "from livelossplot.tf_keras import PlotLossesCallback\n",
    "import tensorflow as tf\n",
    "\n",
    "print(\"Tensorflow Version :\",tf.__version__)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Plot Sample Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Anger: 3995\n",
      "Disgust: 436\n",
      "Fear: 4097\n",
      "Happy: 7215\n",
      "Neutral: 4965\n",
      "Sad: 4830\n",
      "Surprise: 3171\n"
     ]
    }
   ],
   "source": [
    "for expression in os.listdir(\"train/\"):\n",
    "    print(expression+\":\",str(len(os.listdir(\"train/\"+expression))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Generating Train and Validation Batches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 28709 images belonging to 7 classes.\n",
      "Found 7178 images belonging to 7 classes.\n"
     ]
    }
   ],
   "source": [
    "img_size = 48\n",
    "batch_size = 32\n",
    "\n",
    "datagen_train = ImageDataGenerator(horizontal_flip = True,\n",
    "                                   rotation_range = 40,\n",
    "                                  brightness_range=(0.,2.))\n",
    "train_generator = datagen_train.flow_from_directory(\"train/\",\n",
    "                                                   target_size = (img_size,img_size),\n",
    "                                                       color_mode = 'grayscale',\n",
    "                                                        batch_size = batch_size,\n",
    "                                                   class_mode = 'categorical',\n",
    "                                                   shuffle = True)\n",
    "\n",
    "\n",
    "datagen_validation = ImageDataGenerator(horizontal_flip = True,\n",
    "                                   rotation_range = 40,\n",
    "                                  brightness_range=(0.,2.))\n",
    "validation_generator = datagen_validation.flow_from_directory(\"test/\",\n",
    "                                                   target_size = (img_size,img_size),\n",
    "                                                       color_mode = 'grayscale',\n",
    "                                                        batch_size = batch_size,\n",
    "                                                   class_mode = 'categorical',\n",
    "                                                   shuffle = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Define Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_2\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_5 (Conv2D)            (None, 48, 48, 64)        640       \n",
      "_________________________________________________________________\n",
      "batch_normalization_6 (Batch (None, 48, 48, 64)        256       \n",
      "_________________________________________________________________\n",
      "activation_6 (Activation)    (None, 48, 48, 64)        0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2 (None, 24, 24, 64)        0         \n",
      "_________________________________________________________________\n",
      "dropout_6 (Dropout)          (None, 24, 24, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_6 (Conv2D)            (None, 24, 24, 128)       204928    \n",
      "_________________________________________________________________\n",
      "batch_normalization_7 (Batch (None, 24, 24, 128)       512       \n",
      "_________________________________________________________________\n",
      "activation_7 (Activation)    (None, 24, 24, 128)       0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_5 (MaxPooling2 (None, 12, 12, 128)       0         \n",
      "_________________________________________________________________\n",
      "dropout_7 (Dropout)          (None, 12, 12, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_7 (Conv2D)            (None, 12, 12, 512)       590336    \n",
      "_________________________________________________________________\n",
      "batch_normalization_8 (Batch (None, 12, 12, 512)       2048      \n",
      "_________________________________________________________________\n",
      "activation_8 (Activation)    (None, 12, 12, 512)       0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_6 (MaxPooling2 (None, 6, 6, 512)         0         \n",
      "_________________________________________________________________\n",
      "dropout_8 (Dropout)          (None, 6, 6, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 6, 6, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "batch_normalization_9 (Batch (None, 6, 6, 512)         2048      \n",
      "_________________________________________________________________\n",
      "activation_9 (Activation)    (None, 6, 6, 512)         0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_7 (MaxPooling2 (None, 3, 3, 512)         0         \n",
      "_________________________________________________________________\n",
      "dropout_9 (Dropout)          (None, 3, 3, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 4608)              0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 256)               1179904   \n",
      "_________________________________________________________________\n",
      "batch_normalization_10 (Batc (None, 256)               1024      \n",
      "_________________________________________________________________\n",
      "activation_10 (Activation)   (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dropout_10 (Dropout)         (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 512)               131584    \n",
      "_________________________________________________________________\n",
      "batch_normalization_11 (Batc (None, 512)               2048      \n",
      "_________________________________________________________________\n",
      "activation_11 (Activation)   (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dropout_11 (Dropout)         (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 7)                 3591      \n",
      "=================================================================\n",
      "Total params: 4,478,727\n",
      "Trainable params: 4,474,759\n",
      "Non-trainable params: 3,968\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "# 1-conv\n",
    "model.add(Conv2D(64,(3,3),padding='same',input_shape = (48,48,1)))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "# 2-conv\n",
    "model.add(Conv2D(128,(5,5),padding='same'))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "# 3-conv\n",
    "model.add(Conv2D(512,(3,3),padding='same'))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "# 4-conv\n",
    "model.add(Conv2D(512,(3,3),padding='same'))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "\n",
    "model.add(Flatten())\n",
    "\n",
    "model.add(Dense(256))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "model.add(Dense(512))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "model.add(Dense(7,activation='softmax'))\n",
    "\n",
    "opt = Adam(lr=0.0005)\n",
    "\n",
    "model.compile(optimizer=opt,loss='categorical_crossentropy',metrics=['accuracy'])\n",
    "model.summary()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5. Model Training "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Log-loss (cost function):\n",
      "training   (min:    1.116, max:    1.331, cur:    1.119)\n",
      "validation (min:    1.085, max:    1.287, cur:    1.091)\n",
      "\n",
      "accuracy:\n",
      "training   (min:    0.487, max:    0.576, cur:    0.574)\n",
      "validation (min:    0.505, max:    0.592, cur:    0.586)\n",
      "\n",
      "Epoch 00030: saving model to model_weights.h5\n",
      "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
      "897/897 [==============================] - 26s 29ms/step - loss: 1.1188 - accuracy: 0.5742 - val_loss: 1.0911 - val_accuracy: 0.5855\n"
     ]
    }
   ],
   "source": [
    "epochs = 30\n",
    "steps_per_epoch = train_generator.n//train_generator.batch_size\n",
    "validation_steps = validation_generator.n//validation_generator.batch_size\n",
    "\n",
    "checkpoint = ModelCheckpoint(\"model_weights.h5\",monitor='val_accuracy',\n",
    "                            save_weights_only=True, model='max',verbose=1)\n",
    "reduce_lr = ReduceLROnPlateau(monitor='val_loss',factor=0.1,patience=2,min_lr=0.00001,model='auto')\n",
    "\n",
    "callbacks = [PlotLossesCallback(),checkpoint,reduce_lr]\n",
    "history = model.fit(\n",
    "    x = train_generator,\n",
    "    steps_per_epoch=steps_per_epoch,\n",
    "    epochs=epochs,\n",
    "    validation_data=validation_generator,\n",
    "    callbacks=callbacks\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6. Converting model architecture to json file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_json = model.to_json()\n",
    "with open(\"model.json\",\"w\") as json_file:\n",
    "    json_file.write(model_json)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tfgpu21",
   "language": "python",
   "name": "tfgpu21"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}