03_01_autoencoder_train.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Autoencoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "from utils.loaders import load_mnist\n",
    "from models.AE import Autoencoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Set parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# run params\n",
    "SECTION = 'vae'\n",
    "RUN_ID = '0001'\n",
    "DATA_NAME = 'digits'\n",
    "RUN_FOLDER = 'run/{}/'.format(SECTION)\n",
    "RUN_FOLDER += '_'.join([RUN_ID, DATA_NAME])\n",
    "\n",
    "if not os.path.exists(RUN_FOLDER):\n",
    "    os.mkdir(RUN_FOLDER)\n",
    "    os.mkdir(os.path.join(RUN_FOLDER, 'viz'))\n",
    "    os.mkdir(os.path.join(RUN_FOLDER, 'images'))\n",
    "    os.mkdir(os.path.join(RUN_FOLDER, 'weights'))\n",
    "\n",
    "MODE =  'build' #'load' #"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(x_train, y_train), (x_test, y_test) = load_mnist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define the structure of the neural network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "AE = Autoencoder(\n",
    "    input_dim = (28,28,1)\n",
    "    , encoder_conv_filters = [32,64,64, 64]\n",
    "    , encoder_conv_kernel_size = [3,3,3,3]\n",
    "    , encoder_conv_strides = [1,2,2,1]\n",
    "    , decoder_conv_t_filters = [64,64,32,1]\n",
    "    , decoder_conv_t_kernel_size = [3,3,3,3]\n",
    "    , decoder_conv_t_strides = [1,2,2,1]\n",
    "    , z_dim = 2\n",
    ")\n",
    "\n",
    "if MODE == 'build':\n",
    "    AE.save(RUN_FOLDER)\n",
    "else:\n",
    "    AE.load_weights(os.path.join(RUN_FOLDER, 'weights/weights.h5'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "AE.encoder.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "AE.decoder.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train the autoencoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "LEARNING_RATE = 0.0005\n",
    "BATCH_SIZE = 32\n",
    "INITIAL_EPOCH = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "AE.compile(LEARNING_RATE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "AE.train(     \n",
    "    x_train[:1000]\n",
    "    , batch_size = BATCH_SIZE\n",
    "    , epochs = 200\n",
    "    , run_folder = RUN_FOLDER\n",
    "    , initial_epoch = INITIAL_EPOCH\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "gdl_code_2",
   "language": "python",
   "name": "gdl_code_2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}