transfer cat dog (InceptionV3).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# *** Transfer learning using Inception V3 ***"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import applications\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "#from keras.applications import InceptionResNetV2\n",
    "from keras.applications.inception_v3 import InceptionV3\n",
    "from keras import optimizers\n",
    "from keras.models import Sequential, Model \n",
    "from keras.layers import Dropout, Flatten, Dense, GlobalAveragePooling2D\n",
    "from keras import backend as k \n",
    "from keras.callbacks import ModelCheckpoint, LearningRateScheduler, TensorBoard, EarlyStopping"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Initialisng parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "img_width, img_height = 139, 139 # the image dimensions supported by InceptionResNetV2 and InceptionV3, this saves space on GPU\n",
    "train_data_dir = \"data/train\" # the training directory \n",
    "validation_data_dir = \"data/test\" # the directory we will use for validation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Setting the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#model = InceptionResNetV2(weights='imagenet', include_top=False, input_shape = (img_width, img_height, 3))\n",
    "model = InceptionV3(weights='imagenet', include_top=False, input_shape = (img_width, img_height, 3))\n",
    "# we maxed out the resolution for the model to gain the most information\n",
    "# we put a 3 indicating we are taking in the imaages in RGB(colour)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.summary() # seeing the sctructure of the neural network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## if you have layers you don't want you can ignore them\n",
    "# Freeze the layers which you don't want to train. Here I am freezing the first 5 layers.\n",
    "for layer in model.layers[:5]:\n",
    "    layer.trainable = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "##model.summary() # seeing the sctructure of the neural network"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ___Adding layers onto the pre-existing architecture___"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Adding custom Layers \n",
    "x = model.output\n",
    "x = Flatten()(x)\n",
    "x = Dense(64, activation=\"relu\")(x)\n",
    "x = Dropout(0.5)(x)\n",
    "x = Dense(32, activation=\"relu\")(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Defining the fully connected layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "predictions = Dense(units = 1, activation=\"sigmoid\")(x) # sigmoid is for yes or no training oprions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Defining the model for training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating the final model for computation\n",
    "model_final = Model(inputs = model.input, outputs = predictions)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compiling the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# compile the model \n",
    "model_final.compile(loss = 'binary_crossentropy', optimizer = \"Adamax\", metrics=[\"accuracy\"])\n",
    "#Adamax is a variant of Adam optimizer but l prefer the better learning rate it has and...\n",
    "# it is lighter for computation in comparison with Adadelta"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ___Data preparation___"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initiate the training data generator with data Augumentation \n",
    "train_datagen = ImageDataGenerator(\n",
    "rescale = 1./255,\n",
    "horizontal_flip = True,\n",
    "fill_mode = \"nearest\",\n",
    "zoom_range = 0.3,\n",
    "width_shift_range = 0.3,\n",
    "height_shift_range=0.3,\n",
    "rotation_range=30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Initiate the test data generator with data Augumentation \n",
    "test_datagen = ImageDataGenerator(\n",
    "rescale = 1./255,\n",
    "horizontal_flip = True,\n",
    "fill_mode = \"nearest\",\n",
    "zoom_range = 0.3,\n",
    "width_shift_range = 0.3,\n",
    "height_shift_range=0.3,\n",
    "rotation_range=30)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ___Initialising the model for training___"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## number of images to take as batches to train\n",
    "batch_size = 1 #one image is easier is process allowing resources to be better utilized at thhe expense of time though"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_generator = train_datagen.flow_from_directory(\n",
    "train_data_dir,\n",
    "target_size = (img_height, img_width),\n",
    "batch_size = batch_size, \n",
    "class_mode = \"binary\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "validation_generator = test_datagen.flow_from_directory(\n",
    "validation_data_dir,\n",
    "target_size = (img_height, img_width),\n",
    "class_mode = \"binary\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ___Parameters for saving the model___"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Save the model according to the conditions  \n",
    "checkpoint = ModelCheckpoint(\"cat_dog_model.h5\", monitor='val_acc', verbose=1, save_best_only=True, save_weights_only=False, mode='auto', period=1)\n",
    "early = EarlyStopping(monitor='val_acc', min_delta=0, patience=10, verbose=1, mode='auto')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ___Training the model___"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# the number of epochs we want to train our data \n",
    "epochs = 5\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train the model\n",
    "\n",
    "# steps_per_epoch - It should typically be equal to the number of samples of your dataset divided by the batch size. \n",
    "# validation_steps - It should typically be equal to the number of samples of your validation dataset divided by the batch size. \n",
    "model_final.fit_generator(\n",
    "train_generator,\n",
    "steps_per_epoch = 1000,\n",
    "epochs = epochs,\n",
    "verbose=1,\n",
    "validation_data = validation_generator,\n",
    "callbacks = [checkpoint, early],\n",
    "validation_steps = 250)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print ('Done training and saving model \\n Thank you boss')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_____ Done _____\n",
    "<br>\n",
    "https://github.com/simbaTmotsi/cats_vs._Dogs_CNN---training for data allocation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}