Simple-Autoencoder-for-PCA-Udemy-Git-SA.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create Some Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import make_blobs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = make_blobs(n_samples=100, n_features=3,centers=2,random_state=101)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Scale the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = MinMaxScaler()\n",
    "scaled_data = scaler.fit_transform(data[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data[0] \n",
    "data_x = scaled_data[:,0]\n",
    "data_y = scaled_data[:,1]\n",
    "data_z = scaled_data[:,2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The Linear Autoencoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.contrib.layers import fully_connected"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_inputs = 3  # 3 dimensional input\n",
    "num_hidden = 2  # 2 dimensional representation \n",
    "num_outputs = num_inputs # Must be true for an autoencoder!\n",
    "\n",
    "learning_rate = 0.01"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Placeholder\n",
    "\n",
    "Notice there is no real label here, just X."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = tf.placeholder(tf.float32, shape=[None, num_inputs])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Layers\n",
    "\n",
    "Using the fully_connected layers API, we **do not** provide an activation function!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "hidden = fully_connected(X, num_hidden, activation_fn=None)\n",
    "outputs = fully_connected(hidden, num_outputs, activation_fn=None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loss Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss = tf.reduce_mean(tf.square(outputs - X))  # MSE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "optimizer = tf.train.AdamOptimizer(learning_rate)\n",
    "train  = optimizer.minimize( loss)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Init"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Running the Session"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_steps = 1000\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    sess.run(init)\n",
    "    \n",
    "    for iteration in range(num_steps):\n",
    "        sess.run(train,feed_dict={X: scaled_data})\n",
    "\n",
    "        \n",
    "    # Now ask for the hidden layer output (the 2 dimensional output)\n",
    "    output_2d = hidden.eval(feed_dict={X: scaled_data})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100, 2)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output_2d.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x11fcb1e48>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(output_2d[:,0],output_2d[:,1],c=data[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Great Job!"
   ]
  }
 ],
 "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}