00-Generative-Adversarial-Networks.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "___\n",
    "\n",
    "<a href='http://www.pieriandata.com'><img src='../Pierian_Data_Logo.png'/></a>\n",
    "___\n",
    "<center><em>Copyright Pierian Data</em></center>\n",
    "<center><em>For more information, visit us at <a href='http://www.pieriandata.com'>www.pieriandata.com</a></em></center>\n",
    "\n",
    "# GANs - Generative Adverserial Networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.datasets import mnist"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "(X_train, y_train), (X_test, y_test) = mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1d8d93ca108>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.imshow(X_train[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5, 0, 4, ..., 5, 6, 8], dtype=uint8)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Filtering out the Data for Faster Training on Smaller Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "only_zeros = X_train[y_train==0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5923, 28, 28)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "only_zeros.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1d8d9482508>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.imshow(only_zeros[950])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras.layers import Dense,Reshape,Flatten\n",
    "from tensorflow.keras.models import Sequential"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(42)\n",
    "tf.random.set_seed(42)\n",
    "\n",
    "codings_size = 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "generator = Sequential()\n",
    "generator.add(Dense(100, activation=\"relu\", input_shape=[codings_size]))\n",
    "generator.add(Dense(150,activation='relu'))\n",
    "generator.add(Dense(784, activation=\"sigmoid\")) # 28*28 = 784\n",
    "generator.add(Reshape([28,28]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "discriminator = Sequential()\n",
    "discriminator.add(Flatten(input_shape=[28,28]))\n",
    "discriminator.add(Dense(150,activation='relu'))\n",
    "discriminator.add(Dense(100,activation='relu'))\n",
    "discriminator.add(Dense(1,activation=\"sigmoid\"))\n",
    "\n",
    "discriminator.compile(loss=\"binary_crossentropy\", optimizer=\"adam\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "GAN = Sequential([generator, discriminator])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "discriminator.trainable = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "GAN.compile(loss=\"binary_crossentropy\", optimizer=\"adam\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<tensorflow.python.keras.engine.sequential.Sequential at 0x1d8d99390c8>,\n",
       " <tensorflow.python.keras.engine.sequential.Sequential at 0x1d8d9a48908>]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "GAN.layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_2\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "sequential (Sequential)      (None, 28, 28)            143634    \n",
      "_________________________________________________________________\n",
      "sequential_1 (Sequential)    (None, 1)                 132951    \n",
      "=================================================================\n",
      "Total params: 276,585\n",
      "Trainable params: 143,634\n",
      "Non-trainable params: 132,951\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "GAN.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense (Dense)                (None, 100)               10100     \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 150)               15150     \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 784)               118384    \n",
      "_________________________________________________________________\n",
      "reshape (Reshape)            (None, 28, 28)            0         \n",
      "=================================================================\n",
      "Total params: 143,634\n",
      "Trainable params: 143,634\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "GAN.layers[0].summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "flatten (Flatten)            (None, 784)               0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 150)               117750    \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 100)               15100     \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 1)                 101       \n",
      "=================================================================\n",
      "WARNING:tensorflow:Discrepancy between trainable weights and collected trainable weights, did you set `model.trainable` without calling `model.compile` after ?\n",
      "Total params: 265,902\n",
      "Trainable params: 132,951\n",
      "Non-trainable params: 132,951\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "GAN.layers[1].summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Setting up Training Batches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_size = 32"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://stackoverflow.com/questions/46444018/meaning-of-buffer-size-in-dataset-map-dataset-prefetch-and-dataset-shuffle\n",
    "\n",
    "The buffer_size in Dataset.shuffle() can affect the randomness of your dataset, and hence the order in which elements are produced. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# my_data = X_train\n",
    "my_data = only_zeros"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = tf.data.Dataset.from_tensor_slices(my_data).shuffle(buffer_size=1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensorflow.python.data.ops.dataset_ops.ShuffleDataset"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = dataset.batch(batch_size, drop_remainder=True).prefetch(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "epochs = 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**NOTE: The generator never actually sees any real images. It learns by viewing the gradients going back through the discriminator. The better the discrimnator gets through training, the more information the discriminator contains in its gradients, which means the generator can being to make progress in learning how to generate fake images, in our case, fake zeros.**\n",
    "\n",
    "## Training Loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Currently on Epoch 1\n",
      "\tCurrently on batch number 100 of 185\n",
      "TRAINING COMPLETE\n"
     ]
    }
   ],
   "source": [
    "# Grab the seprate components\n",
    "generator, discriminator = GAN.layers\n",
    "\n",
    "# For every epcoh\n",
    "for epoch in range(epochs):\n",
    "    print(f\"Currently on Epoch {epoch+1}\")\n",
    "    i = 0\n",
    "    # For every batch in the dataset\n",
    "    for X_batch in dataset:\n",
    "        i=i+1\n",
    "        if i%100 == 0:\n",
    "            print(f\"\\tCurrently on batch number {i} of {len(my_data)//batch_size}\")\n",
    "        #####################################\n",
    "        ## TRAINING THE DISCRIMINATOR ######\n",
    "        ###################################\n",
    "        \n",
    "        # Create Noise\n",
    "        noise = tf.random.normal(shape=[batch_size, codings_size])\n",
    "        \n",
    "        # Generate numbers based just on noise input\n",
    "        gen_images = generator(noise)\n",
    "        \n",
    "        # Concatenate Generated Images against the Real Ones\n",
    "        # TO use tf.concat, the data types must match!\n",
    "        X_fake_vs_real = tf.concat([gen_images, tf.dtypes.cast(X_batch,tf.float32)], axis=0)\n",
    "        \n",
    "        # Targets set to zero for fake images and 1 for real images\n",
    "        y1 = tf.constant([[0.]] * batch_size + [[1.]] * batch_size)\n",
    "        \n",
    "        # This gets rid of a Keras warning\n",
    "        discriminator.trainable = True\n",
    "        \n",
    "        # Train the discriminator on this batch\n",
    "        discriminator.train_on_batch(X_fake_vs_real, y1)\n",
    "        \n",
    "        \n",
    "        #####################################\n",
    "        ## TRAINING THE GENERATOR     ######\n",
    "        ###################################\n",
    "        \n",
    "        # Create some noise\n",
    "        noise = tf.random.normal(shape=[batch_size, codings_size])\n",
    "        \n",
    "        # We want discriminator to belive that fake images are real\n",
    "        y2 = tf.constant([[1.]] * batch_size)\n",
    "        \n",
    "        # Avois a warning\n",
    "        discriminator.trainable = False\n",
    "        \n",
    "        GAN.train_on_batch(noise, y2)\n",
    "        \n",
    "print(\"TRAINING COMPLETE\")            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Most likely your generator will only learn to create one type of noisey zero\n",
    "# Regardless of what noise is passed in."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://www.quora.com/What-does-it-mean-if-all-produced-images-of-a-GAN-look-the-same"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "noise = tf.random.normal(shape=[10, codings_size])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TensorShape([10, 100])"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "noise.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1d8dada1208>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.imshow(noise)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "image = generator(noise)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1d8dadeeac8>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.imshow(image[5])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Review our conclusion video to learn about what is state-of-the-art in this field! We need very powerful GPUs to achieve realistic results!**"
   ]
  }
 ],
 "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
}