GAN_mnist.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ[\"KERAS_BACKEND\"] = \"tensorflow\"\n",
    "import numpy as np\n",
    "from tqdm import tqdm\n",
    "from tqdm import tqdm_notebook\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kmura\\AppData\\Local\\Programs\\Python\\Python35\\lib\\site-packages\\h5py\\__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n",
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\kmura\\AppData\\Local\\Programs\\Python\\Python35\\lib\\site-packages\\tensorflow\\contrib\\learn\\python\\learn\\datasets\\base.py:198: retry (from tensorflow.contrib.learn.python.learn.datasets.base) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use the retry module or similar alternatives.\n"
     ]
    }
   ],
   "source": [
    "from keras.layers import Input\n",
    "from keras.models import Model, Sequential\n",
    "from keras.layers.core import Reshape, Dense, Dropout, Flatten\n",
    "from keras.layers.advanced_activations import LeakyReLU\n",
    "from keras.layers.convolutional import Conv2D, UpSampling2D\n",
    "from keras.datasets import mnist\n",
    "from keras.optimizers import Adam\n",
    "from keras import backend as K\n",
    "from keras import initializers\n",
    "\n",
    "K.set_image_dim_ordering('th')\n",
    "\n",
    "# Deterministic output.\n",
    "# Tired of seeing the same results every time? Remove the line below.\n",
    "np.random.seed(1000)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# The results are a little better when the dimensionality of the random vector is only 10.\n",
    "# The dimensionality has been left at 100 for consistency with other GAN implementations.\n",
    "randomDim = 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Load MNIST data\n",
    "(X_train, y_train), (X_test, y_test) = mnist.load_data()\n",
    "X_train = (X_train.astype(np.float32) - 127.5)/127.5\n",
    "X_train = X_train[:, np.newaxis, :, :]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Optimizer\n",
    "adam = Adam(lr=0.0002, beta_1=0.5)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generator\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\kmura\\AppData\\Local\\Programs\\Python\\Python35\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:1290: calling reduce_mean (from tensorflow.python.ops.math_ops) with keep_dims is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "keep_dims is deprecated, use keepdims instead\n"
     ]
    }
   ],
   "source": [
    "generator = Sequential()\n",
    "generator.add(Dense(128*7*7, input_dim=randomDim, kernel_initializer=initializers.RandomNormal(stddev=0.02)))\n",
    "generator.add(LeakyReLU(0.2))\n",
    "generator.add(Reshape((128, 7, 7)))\n",
    "generator.add(UpSampling2D(size=(2, 2)))\n",
    "generator.add(Conv2D(64, kernel_size=(5, 5), padding='same'))\n",
    "generator.add(LeakyReLU(0.2))\n",
    "generator.add(UpSampling2D(size=(2, 2)))\n",
    "generator.add(Conv2D(1, kernel_size=(5, 5), padding='same', activation='tanh'))\n",
    "generator.compile(loss='binary_crossentropy', optimizer=adam)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.utils import plot_model\n",
    "\n",
    "plot_model(generator,to_file='generator_demo.png',show_shapes=True)\n",
    "from IPython.display import Image\n",
    "Image(\"generator_demo.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Discriminator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\kmura\\AppData\\Local\\Programs\\Python\\Python35\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:1205: calling reduce_prod (from tensorflow.python.ops.math_ops) with keep_dims is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "keep_dims is deprecated, use keepdims instead\n"
     ]
    }
   ],
   "source": [
    "\n",
    "discriminator = Sequential()\n",
    "discriminator.add(Conv2D(64, kernel_size=(5, 5), strides=(2, 2), padding='same', input_shape=(1, 28, 28), kernel_initializer=initializers.RandomNormal(stddev=0.02)))\n",
    "discriminator.add(LeakyReLU(0.2))\n",
    "discriminator.add(Dropout(0.3))\n",
    "discriminator.add(Conv2D(128, kernel_size=(5, 5), strides=(2, 2), padding='same'))\n",
    "discriminator.add(LeakyReLU(0.2))\n",
    "discriminator.add(Dropout(0.3))\n",
    "discriminator.add(Flatten())\n",
    "discriminator.add(Dense(1, activation='sigmoid'))\n",
    "discriminator.compile(loss='binary_crossentropy', optimizer=adam)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.utils import plot_model\n",
    "\n",
    "plot_model(discriminator,to_file='discriminator_demo.png',show_shapes=True)\n",
    "from IPython.display import Image\n",
    "Image(\"discriminator_demo.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Combined network\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "discriminator.trainable = False\n",
    "ganInput = Input(shape=(randomDim,))\n",
    "x = generator(ganInput)\n",
    "ganOutput = discriminator(x)\n",
    "gan = Model(inputs=ganInput, outputs=ganOutput)\n",
    "gan.compile(loss='binary_crossentropy', optimizer=adam)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.utils import plot_model\n",
    "\n",
    "plot_model(gan,to_file='gan_demo.png',show_shapes=True)\n",
    "from IPython.display import Image\n",
    "Image(\"gan_demo.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dLosses = []\n",
    "gLosses = []\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plot the loss from each batch\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def plotLoss(epoch):\n",
    "    plt.figure(figsize=(10, 8))\n",
    "    plt.plot(dLosses, label='Discriminitive loss')\n",
    "    plt.plot(gLosses, label='Generative loss')\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.ylabel('Loss')\n",
    "    plt.legend()\n",
    "    plt.savefig('images/dcgan_loss_epoch_%d.png' % epoch)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create a wall of generated MNIST images\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def plotGeneratedImages(epoch, examples=100, dim=(10, 10), figsize=(10, 10)):\n",
    "    noise = np.random.normal(0, 1, size=[examples, randomDim])\n",
    "    generatedImages = generator.predict(noise)\n",
    "\n",
    "    plt.figure(figsize=figsize)\n",
    "    for i in range(generatedImages.shape[0]):\n",
    "        plt.subplot(dim[0], dim[1], i+1)\n",
    "        plt.imshow(generatedImages[i, 0], interpolation='nearest', cmap='gray_r')\n",
    "        plt.axis('off')\n",
    "    plt.tight_layout()\n",
    "    plt.savefig('images/dcgan_generated_image_epoch_%d.png' % epoch)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Save the generator and discriminator networks (and weights) for later use\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def saveModels(epoch):\n",
    "    generator.save('models/dcgan_generator_epoch_%d.h5' % epoch)\n",
    "    discriminator.save('models/dcgan_discriminator_epoch_%d.h5' % epoch)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def train(epochs=1, batchSize=128):\n",
    "    batchCount = X_train.shape[0] // batchSize\n",
    "    print ('Epochs:', epochs)\n",
    "    print ('Batch size:', batchSize)\n",
    "    print ('Batches per epoch:', batchCount)\n",
    "\n",
    "    for e in range(1, epochs+1):\n",
    "        print ('-'*15, 'Epoch %d' % e, '-'*15)\n",
    "        print(batchCount)\n",
    "        for _ in range(batchCount):\n",
    "            # Get a random set of input noise and images\n",
    "            noise = np.random.normal(0, 1, size=[batchSize, randomDim])\n",
    "            imageBatch = X_train[np.random.randint(0, X_train.shape[0], size=batchSize)]\n",
    "\n",
    "            # Generate fake MNIST images\n",
    "            generatedImages = generator.predict(noise)\n",
    "            X = np.concatenate([imageBatch, generatedImages])\n",
    "\n",
    "            # Labels for generated and real data\n",
    "            yDis = np.zeros(2*batchSize)\n",
    "            # One-sided label smoothing\n",
    "            yDis[:batchSize] = 0.9\n",
    "\n",
    "            # Train discriminator\n",
    "            discriminator.trainable = True\n",
    "            dloss = discriminator.train_on_batch(X, yDis)\n",
    "\n",
    "            # Train generator\n",
    "            noise = np.random.normal(0, 1, size=[batchSize, randomDim])\n",
    "            yGen = np.ones(batchSize)\n",
    "            discriminator.trainable = False\n",
    "            gloss = gan.train_on_batch(noise, yGen)\n",
    "\n",
    "        # Store loss of most recent batch from this epoch\n",
    "        dLosses.append(dloss)\n",
    "        gLosses.append(gloss)\n",
    "\n",
    "        if e == 1 or e % 5 == 0:\n",
    "            plotGeneratedImages(e)\n",
    "            saveModels(e)\n",
    "\n",
    "    # Plot losses from every epoch\n",
    "    plotLoss(e)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epochs: 50\n",
      "Batch size: 128\n",
      "Batches per epoch: 468\n",
      "--------------- Epoch 1 ---------------\n",
      "468\n",
      "--------------- Epoch 2 ---------------\n",
      "468\n",
      "--------------- Epoch 3 ---------------\n",
      "468\n",
      "--------------- Epoch 4 ---------------\n",
      "468\n",
      "--------------- Epoch 5 ---------------\n",
      "468\n",
      "--------------- Epoch 6 ---------------\n",
      "468\n",
      "--------------- Epoch 7 ---------------\n",
      "468\n",
      "--------------- Epoch 8 ---------------\n",
      "468\n",
      "--------------- Epoch 9 ---------------\n",
      "468\n",
      "--------------- Epoch 10 ---------------\n",
      "468\n",
      "--------------- Epoch 11 ---------------\n",
      "468\n",
      "--------------- Epoch 12 ---------------\n",
      "468\n",
      "--------------- Epoch 13 ---------------\n",
      "468\n",
      "--------------- Epoch 14 ---------------\n",
      "468\n",
      "--------------- Epoch 15 ---------------\n",
      "468\n",
      "--------------- Epoch 16 ---------------\n",
      "468\n",
      "--------------- Epoch 17 ---------------\n",
      "468\n",
      "--------------- Epoch 18 ---------------\n",
      "468\n",
      "--------------- Epoch 19 ---------------\n",
      "468\n",
      "--------------- Epoch 20 ---------------\n",
      "468\n",
      "--------------- Epoch 21 ---------------\n",
      "468\n",
      "--------------- Epoch 22 ---------------\n",
      "468\n",
      "--------------- Epoch 23 ---------------\n",
      "468\n",
      "--------------- Epoch 24 ---------------\n",
      "468\n",
      "--------------- Epoch 25 ---------------\n",
      "468\n",
      "--------------- Epoch 26 ---------------\n",
      "468\n",
      "--------------- Epoch 27 ---------------\n",
      "468\n",
      "--------------- Epoch 28 ---------------\n",
      "468\n",
      "--------------- Epoch 29 ---------------\n",
      "468\n",
      "--------------- Epoch 30 ---------------\n",
      "468\n",
      "--------------- Epoch 31 ---------------\n",
      "468\n",
      "--------------- Epoch 32 ---------------\n",
      "468\n",
      "--------------- Epoch 33 ---------------\n",
      "468\n",
      "--------------- Epoch 34 ---------------\n",
      "468\n",
      "--------------- Epoch 35 ---------------\n",
      "468\n",
      "--------------- Epoch 36 ---------------\n",
      "468\n",
      "--------------- Epoch 37 ---------------\n",
      "468\n",
      "--------------- Epoch 38 ---------------\n",
      "468\n",
      "--------------- Epoch 39 ---------------\n",
      "468\n",
      "--------------- Epoch 40 ---------------\n",
      "468\n",
      "--------------- Epoch 41 ---------------\n",
      "468\n",
      "--------------- Epoch 42 ---------------\n",
      "468\n",
      "--------------- Epoch 43 ---------------\n",
      "468\n",
      "--------------- Epoch 44 ---------------\n",
      "468\n",
      "--------------- Epoch 45 ---------------\n",
      "468\n",
      "--------------- Epoch 46 ---------------\n",
      "468\n",
      "--------------- Epoch 47 ---------------\n",
      "468\n",
      "--------------- Epoch 48 ---------------\n",
      "468\n",
      "--------------- Epoch 49 ---------------\n",
      "468\n",
      "--------------- Epoch 50 ---------------\n",
      "468\n"
     ]
    }
   ],
   "source": [
    "if __name__ == '__main__':\n",
    "    train(50, 128)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "python35",
   "language": "python",
   "name": "python35"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}