Untitled.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0050 cost= 0.150315344 W= 0.401048 b= -0.288088\n",
      "Epoch: 0100 cost= 0.141840130 W= 0.392041 b= -0.223288\n",
      "Epoch: 0150 cost= 0.134343892 W= 0.383569 b= -0.162341\n",
      "Epoch: 0200 cost= 0.127713427 W= 0.375601 b= -0.105019\n",
      "Epoch: 0250 cost= 0.121848904 W= 0.368106 b= -0.0511056\n",
      "Epoch: 0300 cost= 0.116661832 W= 0.361058 b= -0.000399026\n",
      "Epoch: 0350 cost= 0.112074003 W= 0.354428 b= 0.0472917\n",
      "Epoch: 0400 cost= 0.108016223 W= 0.348193 b= 0.0921459\n",
      "Epoch: 0450 cost= 0.104427248 W= 0.342329 b= 0.134333\n",
      "Epoch: 0500 cost= 0.101252995 W= 0.336814 b= 0.174011\n",
      "Epoch: 0550 cost= 0.098445557 W= 0.331626 b= 0.211328\n",
      "Epoch: 0600 cost= 0.095962554 W= 0.326747 b= 0.246427\n",
      "Epoch: 0650 cost= 0.093766548 W= 0.322159 b= 0.279437\n",
      "Epoch: 0700 cost= 0.091824323 W= 0.317843 b= 0.310485\n",
      "Epoch: 0750 cost= 0.090106666 W= 0.313784 b= 0.339686\n",
      "Epoch: 0800 cost= 0.088587575 W= 0.309966 b= 0.36715\n",
      "Epoch: 0850 cost= 0.087244101 W= 0.306375 b= 0.392982\n",
      "Epoch: 0900 cost= 0.086056001 W= 0.302998 b= 0.417276\n",
      "Epoch: 0950 cost= 0.085005336 W= 0.299822 b= 0.440125\n",
      "Epoch: 1000 cost= 0.084076136 W= 0.296835 b= 0.461616\n",
      "Optimization Finished!\n",
      "Training cost= 0.0840761 W= 0.296835 b= 0.461616 \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\naUjhLiKShhTuIiJp6P8Drobi4oVlZOAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x116d89d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing... (Mean square loss Comparison)\n",
      "Testing cost= 0.0776298\n",
      "Absolute mean square loss difference: 0.00644632\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x116e643c8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from __future__ import print_function\n",
    "\n",
    "import tensorflow as tf\n",
    "import numpy\n",
    "import matplotlib.pyplot as plt\n",
    "rng = numpy.random\n",
    "\n",
    "# Parameters\n",
    "learning_rate = 0.01\n",
    "training_epochs = 1000\n",
    "display_step = 50\n",
    "\n",
    "# Training Data\n",
    "train_X = numpy.asarray([3.3,4.4,5.5,6.71,6.93,4.168,9.779,6.182,7.59,2.167,\n",
    "                         7.042,10.791,5.313,7.997,5.654,9.27,3.1])\n",
    "train_Y = numpy.asarray([1.7,2.76,2.09,3.19,1.694,1.573,3.366,2.596,2.53,1.221,\n",
    "                         2.827,3.465,1.65,2.904,2.42,2.94,1.3])\n",
    "n_samples = train_X.shape[0]\n",
    "\n",
    "# tf Graph Input\n",
    "X = tf.placeholder(\"float\")\n",
    "Y = tf.placeholder(\"float\")\n",
    "\n",
    "# Set model weights\n",
    "W = tf.Variable(rng.randn(), name=\"weight\")\n",
    "b = tf.Variable(rng.randn(), name=\"bias\")\n",
    "\n",
    "# Construct a linear model\n",
    "pred = tf.add(tf.multiply(X, W), b)\n",
    "\n",
    "# Mean squared error\n",
    "cost = tf.reduce_sum(tf.pow(pred-Y, 2))/(2*n_samples)\n",
    "# Gradient descent\n",
    "#  Note, minimize() knows to modify W and b because Variable objects are trainable=True by default\n",
    "optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)\n",
    "\n",
    "# Initialize the variables (i.e. assign their default value)\n",
    "init = tf.global_variables_initializer()\n",
    "\n",
    "# Start training\n",
    "with tf.Session() as sess:\n",
    "\n",
    "    # Run the initializer\n",
    "    sess.run(init)\n",
    "\n",
    "    # Fit all training data\n",
    "    for epoch in range(training_epochs):\n",
    "        for (x, y) in zip(train_X, train_Y):\n",
    "            sess.run(optimizer, feed_dict={X: x, Y: y})\n",
    "\n",
    "        # Display logs per epoch step\n",
    "        if (epoch+1) % display_step == 0:\n",
    "            c = sess.run(cost, feed_dict={X: train_X, Y:train_Y})\n",
    "            print(\"Epoch:\", '%04d' % (epoch+1), \"cost=\", \"{:.9f}\".format(c), \\\n",
    "                \"W=\", sess.run(W), \"b=\", sess.run(b))\n",
    "\n",
    "    print(\"Optimization Finished!\")\n",
    "    training_cost = sess.run(cost, feed_dict={X: train_X, Y: train_Y})\n",
    "    print(\"Training cost=\", training_cost, \"W=\", sess.run(W), \"b=\", sess.run(b), '\\n')\n",
    "\n",
    "    # Graphic display\n",
    "    plt.plot(train_X, train_Y, 'ro', label='Original data')\n",
    "    plt.plot(train_X, sess.run(W) * train_X + sess.run(b), label='Fitted line')\n",
    "    plt.legend()\n",
    "    plt.show()\n",
    "\n",
    "    # Testing example, as requested (Issue #2)\n",
    "    test_X = numpy.asarray([6.83, 4.668, 8.9, 7.91, 5.7, 8.7, 3.1, 2.1])\n",
    "    test_Y = numpy.asarray([1.84, 2.273, 3.2, 2.831, 2.92, 3.24, 1.35, 1.03])\n",
    "\n",
    "    print(\"Testing... (Mean square loss Comparison)\")\n",
    "    testing_cost = sess.run(\n",
    "        tf.reduce_sum(tf.pow(pred - Y, 2)) / (2 * test_X.shape[0]),\n",
    "        feed_dict={X: test_X, Y: test_Y})  # same function as cost above\n",
    "    print(\"Testing cost=\", testing_cost)\n",
    "    print(\"Absolute mean square loss difference:\", abs(\n",
    "        training_cost - testing_cost))\n",
    "\n",
    "    plt.plot(test_X, test_Y, 'bo', label='Testing data')\n",
    "    plt.plot(train_X, sess.run(W) * train_X + sess.run(b), label='Fitted line')\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}