POC image compression for MNIST images.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/tensorflow_core/__init__.py:1473: The name tf.estimator.inputs is deprecated. Please use tf.compat.v1.estimator.inputs instead.\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import tensorflow.keras.layers\n",
    "import tensorflow.keras.models\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow.keras.datasets\n",
    "import tensorflow.keras.optimizers \n",
    "import numpy\n",
    "from timeit import default_timer as timer\n",
    "from keras.preprocessing.image import load_img \n",
    "import warnings \n",
    "from keras.preprocessing.image import img_to_array \n",
    "from keras.preprocessing.image import array_to_img "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_size(np):\n",
    "    return np.size * np.itemsize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/tensorflow_core/python/ops/resource_variable_ops.py:1630: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "If using Keras pass *_constraint arguments to layers.\n"
     ]
    }
   ],
   "source": [
    "x = tensorflow.keras.layers.Input(shape=(784), name=\"encoder_input\")\n",
    "\n",
    "encoder_dense_layer1 = tensorflow.keras.layers.Dense(units=300, name=\"encoder_dense_1\")(x)\n",
    "encoder_activ_layer1 = tensorflow.keras.layers.LeakyReLU(name=\"encoder_leakyrelu_1\")(encoder_dense_layer1)\n",
    "\n",
    "encoder_dense_layer2 = tensorflow.keras.layers.Dense(units=8, name=\"encoder_dense_2\")(encoder_activ_layer1)\n",
    "encoder_output = tensorflow.keras.layers.LeakyReLU(name=\"encoder_output\")(encoder_dense_layer2)\n",
    "\n",
    "encoder = tensorflow.keras.models.Model(x, encoder_output, name=\"encoder_model\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "decoder_input = tensorflow.keras.layers.Input(shape=(8), name=\"decoder_input\")\n",
    "\n",
    "decoder_dense_layer1 = tensorflow.keras.layers.Dense(units=300, name=\"decoder_dense_1\")(decoder_input)\n",
    "decoder_activ_layer1 = tensorflow.keras.layers.LeakyReLU(name=\"decoder_leakyrelu_1\")(decoder_dense_layer1)\n",
    "\n",
    "decoder_dense_layer2 = tensorflow.keras.layers.Dense(units=784, name=\"decoder_dense_2\")(decoder_activ_layer1)\n",
    "decoder_output = tensorflow.keras.layers.LeakyReLU(name=\"decoder_output\")(decoder_dense_layer2)\n",
    "\n",
    "decoder = tensorflow.keras.models.Model(decoder_input, decoder_output, name=\"decoder_model\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "ae_input = tensorflow.keras.layers.Input(shape=(784), name=\"AE_input\")\n",
    "ae_encoder_output = encoder(ae_input)\n",
    "ae_decoder_output = decoder(ae_encoder_output)\n",
    "\n",
    "ae = tensorflow.keras.models.Model(ae_input, ae_decoder_output, name=\"AE\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
      "11493376/11490434 [==============================] - 4s 0us/step\n"
     ]
    }
   ],
   "source": [
    "(x_train_orig, y_train), (x_test_orig, y_test) = tensorflow.keras.datasets.mnist.load_data()\n",
    "x_train_orig = x_train_orig.astype(\"float32\") / 255.0\n",
    "x_test_orig = x_test_orig.astype(\"float32\") / 255.0\n",
    "\n",
    "x_train = numpy.reshape(x_train_orig, newshape=(x_train_orig.shape[0], numpy.prod(x_train_orig.shape[1:])))\n",
    "x_test = numpy.reshape(x_test_orig, newshape=(x_test_orig.shape[0], numpy.prod(x_test_orig.shape[1:])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "ename": "RuntimeError",
     "evalue": "You must compile your model before training/testing. Use `model.compile(optimizer, loss)`.",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mRuntimeError\u001b[0m                              Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-7-9c298b5a3180>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mae\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m120\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m256\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mshuffle\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalidation_data\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)\u001b[0m\n\u001b[1;32m    702\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    703\u001b[0m       \u001b[0;32mraise\u001b[0m \u001b[0mTypeError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'Unrecognized keyword arguments: '\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 704\u001b[0;31m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_assert_compile_was_called\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    705\u001b[0m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_check_call_args\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'fit'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    706\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training.py\u001b[0m in \u001b[0;36m_assert_compile_was_called\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m   2900\u001b[0m     \u001b[0;31m# (i.e. whether the model is built and its inputs/outputs are set).\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2901\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimizer\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2902\u001b[0;31m       raise RuntimeError('You must compile your model before '\n\u001b[0m\u001b[1;32m   2903\u001b[0m                          \u001b[0;34m'training/testing. '\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2904\u001b[0m                          'Use `model.compile(optimizer, loss)`.')\n",
      "\u001b[0;31mRuntimeError\u001b[0m: You must compile your model before training/testing. Use `model.compile(optimizer, loss)`."
     ]
    }
   ],
   "source": [
    "ae.fit(x_train, x_train, epochs=120, batch_size=256, shuffle=True, validation_data=(x_test, x_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "encoded_images = encoder.predict(x_train)\n",
    "decoded_images = decoder.predict(encoded_images)\n",
    "\n",
    "decoded_images_orig = numpy.reshape(decoded_images, newshape=(decoded_images.shape[0], 28, 28))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_image(random_number):\n",
    "    plt.subplot(num_images_to_show, 2, 1)\n",
    "    plt.imshow(x_train_orig[random_number, :, :], cmap=\"gray\")\n",
    "    plt.subplot(num_images_to_show, 2, 2)\n",
    "    plt.imshow(decoded_images_orig[random_number, :, :], cmap=\"gray\")\n",
    "    print(f\"Original =>{get_size(x_train_orig[random_number, :, :])}\")\n",
    "    print(f\"Encoded =>{get_size(encoded_images[random_number])}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_image(encoded_image):\n",
    "    start = timer()\n",
    "    decoded_image = decoder.predict([encoded_image])\n",
    "    decoded_image_orig = numpy.reshape(decoded_image, newshape=(decoded_image.shape[0], 28, 28))\n",
    "    plt.subplot(num_images_to_show, 2, 1)\n",
    "    plt.imshow(decoded_image_orig[0, :, :], cmap=\"gray\")\n",
    "    end = timer()\n",
    "    print(f\"Time taken to decompress => {end - start} secs\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot_image(12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "generate_image([encoded_images[215]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ae.save(\"mnist-numbers\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Environment (conda_tensorflow_p36)",
   "language": "python",
   "name": "conda_tensorflow_p36"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}