vae-gan-128.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow.keras as keras\n",
    "import os\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "import datetime\n",
    "import sys"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.config.experimental.list_physical_devices('GPU')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Input dataset\n",
    "Dataset used is celebA : https://www.kaggle.com/jessicali9530/celeba-dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(128, 128, 3)\n",
      "len dataset :  202599\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "IM_DIM = 128\n",
    "@tf.function\n",
    "def im_preprocessing(im_path) :\n",
    "    im_file = tf.io.read_file(im_path)\n",
    "    im = tf.io.decode_jpeg(im_file)\n",
    "    im = tf.image.convert_image_dtype(im, tf.float32)\n",
    "    #im = tf.image.crop_to_bounding_box(im, 20,0,178,178)\n",
    "    im = tf.image.resize(im, [IM_DIM, IM_DIM])\n",
    "    im = tf.image.random_flip_left_right(im)\n",
    "    return(im)\n",
    "\n",
    "batch_size = 8\n",
    "im_patern = \"./data/img_align_celeba/*.jpg\"\n",
    "files_dataset = tf.data.Dataset.list_files(im_patern)\n",
    "im_dataset = files_dataset.map(im_preprocessing)\n",
    "im_dataset_repeated = im_dataset.repeat()\n",
    "im_dataset_batch = im_dataset_repeated.batch(batch_size)\n",
    "\n",
    "for x in im_dataset.take(1) :\n",
    "    print(x.numpy().shape)\n",
    "    print(\"len dataset : \", tf.data.experimental.cardinality(im_dataset).numpy())\n",
    "    plt.imshow(x)\n",
    "    \n",
    "batch_gen = iter(im_dataset_batch)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Making models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "DEPTH = 32\n",
    "LATENT_DEPTH = 1024\n",
    "K_SIZE = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def sampling(args):\n",
    "    mean, logsigma = args\n",
    "    epsilon = keras.backend.random_normal(shape=keras.backend.shape(mean))\n",
    "    return mean + tf.exp(logsigma / 2) * epsilon\n",
    "\n",
    "def encoder():\n",
    "    input_E = keras.layers.Input(shape=(IM_DIM, IM_DIM, 3))\n",
    "    \n",
    "    X = keras.layers.Conv2D(filters=DEPTH*2, kernel_size=K_SIZE, strides=2, padding='same')(input_E)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "\n",
    "    X = keras.layers.Conv2D(filters=DEPTH*4, kernel_size=K_SIZE, strides=2, padding='same')(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "\n",
    "    X = keras.layers.Conv2D(filters=DEPTH*8, kernel_size=K_SIZE, strides=2, padding='same')(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "\n",
    "    X = keras.layers.Conv2D(filters=DEPTH*8, kernel_size=K_SIZE, strides=2, padding='same')(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "\n",
    "    X = keras.layers.Flatten()(X)\n",
    "    X = keras.layers.Dense(LATENT_DEPTH)(X)    \n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "    \n",
    "    mean = keras.layers.Dense(LATENT_DEPTH,activation=\"tanh\")(X)\n",
    "    logsigma = keras.layers.Dense(LATENT_DEPTH,activation=\"tanh\")(X)\n",
    "    latent = keras.layers.Lambda(sampling, output_shape=(LATENT_DEPTH,))([mean, logsigma])\n",
    "    \n",
    "    kl_loss = 1 + logsigma - keras.backend.square(mean) - keras.backend.exp(logsigma)\n",
    "    kl_loss = keras.backend.mean(kl_loss, axis=-1)\n",
    "    kl_loss *= -0.5\n",
    "    \n",
    "    return keras.models.Model(input_E, [latent,kl_loss])\n",
    "\n",
    "def generator():\n",
    "    input_G = keras.layers.Input(shape=(LATENT_DEPTH,))\n",
    "\n",
    "    X = keras.layers.Dense(IM_DIM//16*IM_DIM//16*DEPTH*16)(input_G)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "    X = keras.layers.Reshape((IM_DIM//16, IM_DIM//16, DEPTH * 16))(X)\n",
    "    \n",
    "    X = keras.layers.Conv2DTranspose(filters=DEPTH*8, kernel_size=K_SIZE, strides=2, padding='same')(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "\n",
    "    X = keras.layers.Conv2DTranspose(filters=DEPTH*8, kernel_size=K_SIZE, strides=2, padding='same')(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "\n",
    "    X = keras.layers.Conv2DTranspose(filters=DEPTH*4, kernel_size=K_SIZE, strides=2, padding='same')(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "    \n",
    "    X = keras.layers.Conv2DTranspose(filters=DEPTH, kernel_size=K_SIZE, strides=2, padding='same')(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "    \n",
    "    X = keras.layers.Conv2D(filters=3, kernel_size=K_SIZE, padding='same')(X)\n",
    "    X = keras.layers.Activation('sigmoid')(X)\n",
    "\n",
    "    return keras.models.Model(input_G, X)\n",
    "\n",
    "def discriminator():\n",
    "    input_D = keras.layers.Input(shape=(IM_DIM, IM_DIM, 3))\n",
    "    \n",
    "    X = keras.layers.Conv2D(filters=DEPTH, kernel_size=K_SIZE, strides=2, padding='same')(input_D)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "    \n",
    "    X = keras.layers.Conv2D(filters=DEPTH*4, kernel_size=K_SIZE, strides=2, padding='same')(input_D)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "\n",
    "    X = keras.layers.Conv2D(filters=DEPTH*8, kernel_size=K_SIZE, strides=2, padding='same')(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "\n",
    "    X = keras.layers.Conv2D(filters=DEPTH*8, kernel_size=K_SIZE, padding='same')(X)\n",
    "    inner_output = keras.layers.Flatten()(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "    \n",
    "    X = keras.layers.Flatten()(X)\n",
    "    X = keras.layers.Dense(DEPTH*8)(X)\n",
    "    X = keras.layers.BatchNormalization()(X)\n",
    "    X = keras.layers.LeakyReLU(alpha=0.2)(X)\n",
    "    \n",
    "    output = keras.layers.Dense(1)(X)    \n",
    "    \n",
    "    return keras.models.Model(input_D, [output, inner_output])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "for x in im_dataset_batch.take(1) :\n",
    "    test_images = x\n",
    "test_r = tf.random.normal((batch_size, LATENT_DEPTH))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#E = encoder()\n",
    "#G = generator()\n",
    "#D = discriminator() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:No training configuration found in save file: the model was *not* compiled. Compile it manually.\n",
      "WARNING:tensorflow:No training configuration found in save file: the model was *not* compiled. Compile it manually.\n",
      "WARNING:tensorflow:No training configuration found in save file: the model was *not* compiled. Compile it manually.\n"
     ]
    }
   ],
   "source": [
    "E = keras.models.load_model(\"saved-models/E_training2_18.h5\")\n",
    "G = keras.models.load_model(\"saved-models/G_training2_18.h5\")\n",
    "D = keras.models.load_model(\"saved-models/D_training2_18.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Step : 60  gan_loss 0.021 vae_loss 0.111 fake_dis_loss 0.011 r_dis_loss 0.003 t_dis_loss 0.035 vae_inner_loss 0.339 E_loss 0.378 D_loss 0.021 kl_loss 0.433 normal_loss 0.353"
     ]
    }
   ],
   "source": [
    "lr=0.00001\n",
    "#lr=0.0001\n",
    "E_opt = keras.optimizers.Adam(lr=lr)\n",
    "G_opt = keras.optimizers.Adam(lr=lr)\n",
    "D_opt = keras.optimizers.Adam(lr=lr)\n",
    "\n",
    "inner_loss_coef = 1\n",
    "normal_coef = 0.1\n",
    "kl_coef = 0.01\n",
    "\n",
    "@tf.function\n",
    "def train_step_vaegan(x):\n",
    "    lattent_r =  tf.random.normal((batch_size, LATENT_DEPTH))\n",
    "    with tf.GradientTape(persistent=True) as tape :\n",
    "        lattent,kl_loss = E(x)\n",
    "        fake = G(lattent)\n",
    "        dis_fake,dis_inner_fake = D(fake)\n",
    "        dis_fake_r,_ = D(G(lattent_r))\n",
    "        dis_true,dis_inner_true = D(x)\n",
    "\n",
    "        vae_inner = dis_inner_fake-dis_inner_true\n",
    "        vae_inner = vae_inner*vae_inner\n",
    "        \n",
    "        mean,var = tf.nn.moments(E(x)[0], axes=0)\n",
    "        var_to_one = var - 1\n",
    "        \n",
    "        normal_loss = tf.reduce_mean(mean*mean) + tf.reduce_mean(var_to_one*var_to_one)\n",
    "        \n",
    "        kl_loss = tf.reduce_mean(kl_loss)\n",
    "        vae_diff_loss = tf.reduce_mean(vae_inner)\n",
    "        f_dis_loss = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(tf.zeros_like(dis_fake), dis_fake))\n",
    "        r_dis_loss = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(tf.zeros_like(dis_fake_r), dis_fake_r))\n",
    "        t_dis_loss = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(tf.ones_like(dis_true), dis_true))\n",
    "        gan_loss = (0.5*t_dis_loss + 0.25*f_dis_loss + 0.25*r_dis_loss)\n",
    "        vae_loss = tf.reduce_mean(tf.abs(x-fake)) \n",
    "        E_loss = vae_diff_loss + kl_coef*kl_loss + normal_coef*normal_loss\n",
    "        G_loss = inner_loss_coef*vae_diff_loss - gan_loss\n",
    "        D_loss = gan_loss\n",
    "    \n",
    "    E_grad = tape.gradient(E_loss,E.trainable_variables)\n",
    "    G_grad = tape.gradient(G_loss,G.trainable_variables)\n",
    "    D_grad = tape.gradient(D_loss,D.trainable_variables)\n",
    "    del tape\n",
    "    E_opt.apply_gradients(zip(E_grad, E.trainable_variables))\n",
    "    G_opt.apply_gradients(zip(G_grad, G.trainable_variables))\n",
    "    D_opt.apply_gradients(zip(D_grad, D.trainable_variables))\n",
    "\n",
    "    return [gan_loss, vae_loss, f_dis_loss, r_dis_loss, t_dis_loss, vae_diff_loss, E_loss, D_loss, kl_loss, normal_loss]\n",
    "\n",
    "step = 0\n",
    "max_step = 10000000\n",
    "log_freq,img_log_freq = 10, 100\n",
    "save_freq,save_number_mult = 1000, 10000\n",
    "\n",
    "metrics_names = [\"gan_loss\", \"vae_loss\", \"fake_dis_loss\", \"r_dis_loss\", \"t_dis_loss\", \"vae_inner_loss\", \"E_loss\", \"D_loss\", \"kl_loss\", \"normal_loss\"]\n",
    "metrics = []\n",
    "for m in metrics_names :\n",
    "    metrics.append(tf.keras.metrics.Mean('m', dtype=tf.float32))\n",
    "\n",
    "current_time = datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\")\n",
    "train_log_dir = ('logs/sep_D%dL%d/' % (DEPTH,LATENT_DEPTH)) + current_time \n",
    "train_summary_writer = tf.summary.create_file_writer(train_log_dir)\n",
    "name = ('sep_D%dL%d' % (DEPTH,LATENT_DEPTH))\n",
    "\n",
    "def save_model() :\n",
    "    nb = str(step // save_number_mult)\n",
    "    D.save('saved-models/D_training3_' + nb + '.h5')\n",
    "    G.save('saved-models/G_training3_' + nb + '.h5')\n",
    "    E.save('saved-models/E_training3_' + nb + '.h5')\n",
    "\n",
    "def print_metrics() :\n",
    "    s = \"\"\n",
    "    for name,metric in zip(metrics_names,metrics) :\n",
    "        s+= \" \" + name + \" \" + str(np.around(metric.result().numpy(), 3)) \n",
    "    print(f\"\\rStep : \" + str(step) + \" \" + s, end=\"\", flush=True)\n",
    "    with train_summary_writer.as_default():\n",
    "        for name,metric in zip(metrics_names,metrics) :\n",
    "            tf.summary.scalar(name, metric.result(), step=step)\n",
    "    for metric in metrics : \n",
    "        metric.reset_states()\n",
    "\n",
    "        \n",
    "def log_images() :\n",
    "    with train_summary_writer.as_default():\n",
    "        lattent,_ = E(test_images)\n",
    "        fake = G(lattent)\n",
    "        fake_r = G(test_r)\n",
    "        tf.summary.image(\"reconstructed image\", fake[:8], step=step, max_outputs=8)\n",
    "        tf.summary.image(\"random image\", fake_r[:8], step=step, max_outputs=8)\n",
    "        dis_fake,inner_dis_fake = D(fake)\n",
    "        dis_fake_r,inner_dis_fake_r = D(fake_r)\n",
    "        dis_true,inner_dis_true = D(test_images)\n",
    "        tf.summary.histogram(\"dis fake\", inner_dis_fake, step=step, buckets=20)\n",
    "        tf.summary.histogram(\"dis true\", inner_dis_true, step=step, buckets=20)\n",
    "        tf.summary.histogram(\"dis random\", inner_dis_fake_r, step=step, buckets=20)\n",
    "        tf.summary.histogram(\"dis lattent\", lattent, step=step, buckets=20)\n",
    "        tf.summary.histogram(\"dis normal\", tf.random.normal((batch_size, LATENT_DEPTH)), step=step, buckets=20)        \n",
    "\n",
    "for x in batch_gen :\n",
    "    step += 1\n",
    "    if not step % log_freq :\n",
    "        print_metrics()\n",
    "    if not step % img_log_freq :\n",
    "        log_images()\n",
    "    if not step % save_freq :\n",
    "        save_model()\n",
    "    \n",
    "    results = train_step_vaegan(x)\n",
    "    for metric,result in zip(metrics, results) :\n",
    "        metric(result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}