stochastic-gradient-ascent.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training Logistic Regression via Stochastic Gradient Ascent\n",
    "\n",
    "The goal of this notebook is to implement a logistic regression classifier using stochastic gradient ascent. We will:\n",
    "\n",
    " * Extract features from Amazon product reviews.\n",
    " * Convert an SFrame into a NumPy array.\n",
    " * Write a function to compute the derivative of log likelihood function with respect to a single coefficient.\n",
    " * Implement stochastic gradient ascent.\n",
    " * Compare convergence of stochastic gradient ascent with that of batch gradient ascent."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fire up GraphLab Create\n",
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from __future__ import division\n",
    "import graphlab"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load and process review dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For this notebook, we will use the same subset of the Amazon product review dataset. The subset was chosen to contain similar numbers of positive and negative reviews, as the original dataset consisted of mostly positive reviews."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[INFO] graphlab.cython.cy_server: GraphLab Create v2.1 started. Logging: /tmp/graphlab_server_1530298665.log\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This non-commercial license of GraphLab Create for academic use is assigned to nanlee_89@yahoo.com and will expire on December 07, 2018.\n"
     ]
    }
   ],
   "source": [
    "products = graphlab.SFrame('amazon_baby_subset.gl/')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will work with a hand-curated list of important words extracted from the review data. We will also perform 2 simple data transformations:\n",
    "\n",
    "1. Remove punctuation using [Python's built-in](https://docs.python.org/2/library/string.html) string manipulation functionality.\n",
    "2. Compute word counts (only for the important_words)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import json\n",
    "with open('important_words.json', 'r') as f: \n",
    "    important_words = json.load(f)\n",
    "important_words = [str(s) for s in important_words]\n",
    "\n",
    "# Remote punctuation\n",
    "def remove_punctuation(text):\n",
    "    import string\n",
    "    return text.translate(None, string.punctuation) \n",
    "\n",
    "products['review_clean'] = products['review'].apply(remove_punctuation)\n",
    "\n",
    "# Split out the words into individual columns\n",
    "for word in important_words:\n",
    "    products[word] = products['review_clean'].apply(lambda s : s.split().count(word))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "The SFrame **products** now contains one column for each of the 193 **important_words**. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\"><table frame=\"box\" rules=\"cols\">\n",
       "    <tr>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">name</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">review</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">rating</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">sentiment</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">review_clean</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">baby</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Stop Pacifier Sucking<br>without tears with ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">All of my kids have cried<br>non-stop when I tried to ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">5.0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">All of my kids have cried<br>nonstop when I tried to ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Nature's Lullabies Second<br>Year Sticker Calendar ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">We wanted to get<br>something to keep track ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">5.0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">We wanted to get<br>something to keep track ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Nature's Lullabies Second<br>Year Sticker Calendar ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">My daughter had her 1st<br>baby over a year ago. ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">5.0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">My daughter had her 1st<br>baby over a year ago She ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Lamaze Peekaboo, I Love<br>You ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">One of baby's first and<br>favorite books, and i ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">4.0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">One of babys first and<br>favorite books and it is ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">SoftPlay Peek-A-Boo<br>Where's Elmo A Childr ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Very cute interactive<br>book! My son loves this ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">5.0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Very cute interactive<br>book My son loves this ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Our Baby Girl Memory Book</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Beautiful book, I love it<br>to record cherished t ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">5.0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Beautiful book I love it<br>to record cherished t ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Hunnt&amp;reg; Falling<br>Flowers and Birds Kids ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Try this out for a spring<br>project !Easy ,fun and ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">5.0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Try this out for a spring<br>project Easy fun and ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Blessed By Pope Benedict<br>XVI Divine Mercy Full ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">very nice Divine Mercy<br>Pendant of Jesus now on ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">5.0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">very nice Divine Mercy<br>Pendant of Jesus now on ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Cloth Diaper Pins<br>Stainless Steel ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">We bought the pins as my<br>6 year old Autistic son ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">4.0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">We bought the pins as my<br>6 year old Autistic son ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">Cloth Diaper Pins<br>Stainless Steel ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">It has been many years<br>since we needed diaper ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">5.0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">It has been many years<br>since we needed diaper ...</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "</table>\n",
       "<table frame=\"box\" rules=\"cols\">\n",
       "    <tr>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">one</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">great</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">love</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">use</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">would</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">like</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">easy</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">little</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">seat</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">old</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">well</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">get</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">also</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">really</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">son</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">time</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">bought</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "</table>\n",
       "<table frame=\"box\" rules=\"cols\">\n",
       "    <tr>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">product</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">good</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">daughter</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">much</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">loves</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">stroller</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">put</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">months</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">car</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">still</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">back</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">used</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">recommend</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">first</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">even</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "    </tr>\n",
       "</table>\n",
       "<table frame=\"box\" rules=\"cols\">\n",
       "    <tr>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">perfect</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">nice</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">...</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">...</td>\n",
       "    </tr>\n",
       "</table>\n",
       "[53072 rows x 198 columns]<br/>Note: Only the head of the SFrame is printed.<br/>You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.\n",
       "</div>"
      ],
      "text/plain": [
       "Columns:\n",
       "\tname\tstr\n",
       "\treview\tstr\n",
       "\trating\tfloat\n",
       "\tsentiment\tint\n",
       "\treview_clean\tstr\n",
       "\tbaby\tint\n",
       "\tone\tint\n",
       "\tgreat\tint\n",
       "\tlove\tint\n",
       "\tuse\tint\n",
       "\twould\tint\n",
       "\tlike\tint\n",
       "\teasy\tint\n",
       "\tlittle\tint\n",
       "\tseat\tint\n",
       "\told\tint\n",
       "\twell\tint\n",
       "\tget\tint\n",
       "\talso\tint\n",
       "\treally\tint\n",
       "\tson\tint\n",
       "\ttime\tint\n",
       "\tbought\tint\n",
       "\tproduct\tint\n",
       "\tgood\tint\n",
       "\tdaughter\tint\n",
       "\tmuch\tint\n",
       "\tloves\tint\n",
       "\tstroller\tint\n",
       "\tput\tint\n",
       "\tmonths\tint\n",
       "\tcar\tint\n",
       "\tstill\tint\n",
       "\tback\tint\n",
       "\tused\tint\n",
       "\trecommend\tint\n",
       "\tfirst\tint\n",
       "\teven\tint\n",
       "\tperfect\tint\n",
       "\tnice\tint\n",
       "\tbag\tint\n",
       "\ttwo\tint\n",
       "\tusing\tint\n",
       "\tgot\tint\n",
       "\tfit\tint\n",
       "\taround\tint\n",
       "\tdiaper\tint\n",
       "\tenough\tint\n",
       "\tmonth\tint\n",
       "\tprice\tint\n",
       "\tgo\tint\n",
       "\tcould\tint\n",
       "\tsoft\tint\n",
       "\tsince\tint\n",
       "\tbuy\tint\n",
       "\troom\tint\n",
       "\tworks\tint\n",
       "\tmade\tint\n",
       "\tchild\tint\n",
       "\tkeep\tint\n",
       "\tsize\tint\n",
       "\tsmall\tint\n",
       "\tneed\tint\n",
       "\tyear\tint\n",
       "\tbig\tint\n",
       "\tmake\tint\n",
       "\ttake\tint\n",
       "\teasily\tint\n",
       "\tthink\tint\n",
       "\tcrib\tint\n",
       "\tclean\tint\n",
       "\tway\tint\n",
       "\tquality\tint\n",
       "\tthing\tint\n",
       "\tbetter\tint\n",
       "\twithout\tint\n",
       "\tset\tint\n",
       "\tnew\tint\n",
       "\tevery\tint\n",
       "\tcute\tint\n",
       "\tbest\tint\n",
       "\tbottles\tint\n",
       "\twork\tint\n",
       "\tpurchased\tint\n",
       "\tright\tint\n",
       "\tlot\tint\n",
       "\tside\tint\n",
       "\thappy\tint\n",
       "\tcomfortable\tint\n",
       "\ttoy\tint\n",
       "\table\tint\n",
       "\tkids\tint\n",
       "\tbit\tint\n",
       "\tnight\tint\n",
       "\tlong\tint\n",
       "\tfits\tint\n",
       "\tsee\tint\n",
       "\tus\tint\n",
       "\tanother\tint\n",
       "\tplay\tint\n",
       "\tday\tint\n",
       "\tmoney\tint\n",
       "\tmonitor\tint\n",
       "\ttried\tint\n",
       "\tthought\tint\n",
       "\tnever\tint\n",
       "\titem\tint\n",
       "\thard\tint\n",
       "\tplastic\tint\n",
       "\thowever\tint\n",
       "\tdisappointed\tint\n",
       "\treviews\tint\n",
       "\tsomething\tint\n",
       "\tgoing\tint\n",
       "\tpump\tint\n",
       "\tbottle\tint\n",
       "\tcup\tint\n",
       "\twaste\tint\n",
       "\treturn\tint\n",
       "\tamazon\tint\n",
       "\tdifferent\tint\n",
       "\ttop\tint\n",
       "\twant\tint\n",
       "\tproblem\tint\n",
       "\tknow\tint\n",
       "\twater\tint\n",
       "\ttry\tint\n",
       "\treceived\tint\n",
       "\tsure\tint\n",
       "\ttimes\tint\n",
       "\tchair\tint\n",
       "\tfind\tint\n",
       "\thold\tint\n",
       "\tgate\tint\n",
       "\topen\tint\n",
       "\tbottom\tint\n",
       "\taway\tint\n",
       "\tactually\tint\n",
       "\tcheap\tint\n",
       "\tworked\tint\n",
       "\tgetting\tint\n",
       "\tordered\tint\n",
       "\tcame\tint\n",
       "\tmilk\tint\n",
       "\tbad\tint\n",
       "\tpart\tint\n",
       "\tworth\tint\n",
       "\tfound\tint\n",
       "\tcover\tint\n",
       "\tmany\tint\n",
       "\tdesign\tint\n",
       "\tlooking\tint\n",
       "\tweeks\tint\n",
       "\tsay\tint\n",
       "\twanted\tint\n",
       "\tlook\tint\n",
       "\tplace\tint\n",
       "\tpurchase\tint\n",
       "\tlooks\tint\n",
       "\tsecond\tint\n",
       "\tpiece\tint\n",
       "\tbox\tint\n",
       "\tpretty\tint\n",
       "\ttrying\tint\n",
       "\tdifficult\tint\n",
       "\ttogether\tint\n",
       "\tthough\tint\n",
       "\tgive\tint\n",
       "\tstarted\tint\n",
       "\tanything\tint\n",
       "\tlast\tint\n",
       "\tcompany\tint\n",
       "\tcome\tint\n",
       "\treturned\tint\n",
       "\tmaybe\tint\n",
       "\ttook\tint\n",
       "\tbroke\tint\n",
       "\tmakes\tint\n",
       "\tstay\tint\n",
       "\tinstead\tint\n",
       "\tidea\tint\n",
       "\thead\tint\n",
       "\tsaid\tint\n",
       "\tless\tint\n",
       "\twent\tint\n",
       "\tworking\tint\n",
       "\thigh\tint\n",
       "\tunit\tint\n",
       "\tseems\tint\n",
       "\tpicture\tint\n",
       "\tcompletely\tint\n",
       "\twish\tint\n",
       "\tbuying\tint\n",
       "\tbabies\tint\n",
       "\twon\tint\n",
       "\ttub\tint\n",
       "\talmost\tint\n",
       "\teither\tint\n",
       "\n",
       "Rows: 53072\n",
       "\n",
       "Data:\n",
       "+-------------------------------+-------------------------------+--------+-----------+\n",
       "|              name             |             review            | rating | sentiment |\n",
       "+-------------------------------+-------------------------------+--------+-----------+\n",
       "| Stop Pacifier Sucking with... | All of my kids have cried ... |  5.0   |     1     |\n",
       "| Nature's Lullabies Second ... | We wanted to get something... |  5.0   |     1     |\n",
       "| Nature's Lullabies Second ... | My daughter had her 1st ba... |  5.0   |     1     |\n",
       "|  Lamaze Peekaboo, I Love You  | One of baby's first and fa... |  4.0   |     1     |\n",
       "| SoftPlay Peek-A-Boo Where'... | Very cute interactive book... |  5.0   |     1     |\n",
       "|   Our Baby Girl Memory Book   | Beautiful book, I love it ... |  5.0   |     1     |\n",
       "| Hunnt&reg; Falling Flowers... | Try this out for a spring ... |  5.0   |     1     |\n",
       "| Blessed By Pope Benedict X... | very nice Divine Mercy Pen... |  5.0   |     1     |\n",
       "| Cloth Diaper Pins Stainles... | We bought the pins as my 6... |  4.0   |     1     |\n",
       "| Cloth Diaper Pins Stainles... | It has been many years sin... |  5.0   |     1     |\n",
       "+-------------------------------+-------------------------------+--------+-----------+\n",
       "+-------------------------------+------+-----+-------+------+-----+-------+------+\n",
       "|          review_clean         | baby | one | great | love | use | would | like |\n",
       "+-------------------------------+------+-----+-------+------+-----+-------+------+\n",
       "| All of my kids have cried ... |  0   |  0  |   1   |  0   |  0  |   0   |  0   |\n",
       "| We wanted to get something... |  0   |  0  |   0   |  0   |  0  |   0   |  0   |\n",
       "| My daughter had her 1st ba... |  1   |  0  |   0   |  0   |  0  |   0   |  0   |\n",
       "| One of babys first and fav... |  0   |  0  |   0   |  0   |  0  |   0   |  1   |\n",
       "| Very cute interactive book... |  0   |  0  |   1   |  0   |  0  |   0   |  0   |\n",
       "| Beautiful book I love it t... |  0   |  0  |   1   |  1   |  0  |   0   |  0   |\n",
       "| Try this out for a spring ... |  0   |  0  |   0   |  0   |  0  |   0   |  0   |\n",
       "| very nice Divine Mercy Pen... |  0   |  0  |   0   |  0   |  0  |   0   |  0   |\n",
       "| We bought the pins as my 6... |  0   |  1  |   0   |  0   |  1  |   0   |  0   |\n",
       "| It has been many years sin... |  0   |  1  |   0   |  0   |  0  |   0   |  1   |\n",
       "+-------------------------------+------+-----+-------+------+-----+-------+------+\n",
       "+------+--------+------+-----+------+-----+------+--------+-----+\n",
       "| easy | little | seat | old | well | get | also | really | ... |\n",
       "+------+--------+------+-----+------+-----+------+--------+-----+\n",
       "|  1   |   0    |  0   |  0  |  0   |  0  |  0   |   0    | ... |\n",
       "|  0   |   0    |  0   |  0  |  0   |  1  |  0   |   0    | ... |\n",
       "|  0   |   0    |  0   |  1  |  0   |  1  |  0   |   0    | ... |\n",
       "|  0   |   0    |  0   |  0  |  0   |  0  |  0   |   0    | ... |\n",
       "|  0   |   0    |  0   |  0  |  0   |  0  |  0   |   0    | ... |\n",
       "|  0   |   0    |  0   |  0  |  0   |  0  |  0   |   0    | ... |\n",
       "|  0   |   0    |  0   |  0  |  0   |  0  |  0   |   0    | ... |\n",
       "|  0   |   0    |  0   |  0  |  0   |  0  |  0   |   0    | ... |\n",
       "|  0   |   0    |  0   |  1  |  0   |  0  |  0   |   0    | ... |\n",
       "|  0   |   1    |  0   |  0  |  0   |  1  |  0   |   0    | ... |\n",
       "+------+--------+------+-----+------+-----+------+--------+-----+\n",
       "[53072 rows x 198 columns]\n",
       "Note: Only the head of the SFrame is printed.\n",
       "You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns."
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "products"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Split data into training and validation sets\n",
    "\n",
    "We will now split the data into a 90-10 split where 90% is in the training set and 10% is in the validation set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set  : 47780 data points\n",
      "Validation set: 5292 data points\n"
     ]
    }
   ],
   "source": [
    "train_data, validation_data = products.random_split(.9, seed=1)\n",
    "\n",
    "print 'Training set  : %d data points' % len(train_data)\n",
    "print 'Validation set: %d data points' % len(validation_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Convert SFrame to NumPy array\n",
    "\n",
    "The following function extracts columns from an SFrame and converts them into a NumPy array. Two arrays are returned: one representing features and another representing class labels. \n",
    "\n",
    "**Note:** The feature matrix includes an additional column 'intercept' filled with 1's to take account of the intercept term."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "def get_numpy_data(data_sframe, features, label):\n",
    "    data_sframe['intercept'] = 1\n",
    "    features = ['intercept'] + features\n",
    "    features_sframe = data_sframe[features]\n",
    "    feature_matrix = features_sframe.to_numpy()\n",
    "    label_sarray = data_sframe[label]\n",
    "    label_array = label_sarray.to_numpy()\n",
    "    return(feature_matrix, label_array)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that we convert both the training and validation sets into NumPy arrays."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "feature_matrix_train, sentiment_train = get_numpy_data(train_data, important_words, 'sentiment')\n",
    "feature_matrix_valid, sentiment_valid = get_numpy_data(validation_data, important_words, 'sentiment') "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building on logistic regression\n",
    "\n",
    "Let us now build on Module 3 assignment. Recall that the link function for logistic regression can be defined as:\n",
    "\n",
    "$$\n",
    "P(y_i = +1 | \\mathbf{x}_i,\\mathbf{w}) = \\frac{1}{1 + \\exp(-\\mathbf{w}^T h(\\mathbf{x}_i))},\n",
    "$$\n",
    "\n",
    "where the feature vector $h(\\mathbf{x}_i)$ is given by the word counts of **important_words** in the review $\\mathbf{x}_i$. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "'''\n",
    "produces probablistic estimate for P(y_i = +1 | x_i, w).\n",
    "estimate ranges between 0 and 1.\n",
    "'''\n",
    "def predict_probability(feature_matrix, coefficients):\n",
    "    # Take dot product of feature_matrix and coefficients  \n",
    "    score = np.dot(feature_matrix, coefficients)\n",
    "    \n",
    "    # Compute P(y_i = +1 | x_i, w) using the link function\n",
    "    predictions = 1. / (1.+np.exp(-score))    \n",
    "    return predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Derivative of log likelihood with respect to a single coefficient\n",
    "\n",
    "Let us now work on making minor changes to how the derivative computation is performed for logistic regression.\n",
    "\n",
    "**The derivative of log likelihood with respect to a single coefficient** is as follows:\n",
    "\n",
    "$$\n",
    "\\frac{\\partial\\ell}{\\partial w_j} = \\sum_{i=1}^N h_j(\\mathbf{x}_i)\\left(\\mathbf{1}[y_i = +1] - P(y_i = +1 | \\mathbf{x}_i, \\mathbf{w})\\right)\n",
    "$$\n",
    "\n",
    "In previous topics, we wrote a function to compute the derivative of log likelihood with respect to a single coefficient $w_j$. The function accepts the following two parameters:\n",
    " * `errors` vector containing $(\\mathbf{1}[y_i = +1] - P(y_i = +1 | \\mathbf{x}_i, \\mathbf{w}))$ for all $i$\n",
    " * `feature` vector containing $h_j(\\mathbf{x}_i)$  for all $i$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def feature_derivative(errors, feature): \n",
    "    \n",
    "    # Compute the dot product of errors and feature\n",
    "    derivative = np.dot(errors, feature)\n",
    "\n",
    "    return derivative"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To verify the correctness of the gradient computation, we write a function for computing average log likelihood.\n",
    "\n",
    "To track the performance of stochastic gradient ascent, we provide a function for computing **average log likelihood**. \n",
    "\n",
    "$$\\ell\\ell_A(\\mathbf{w}) = \\color{red}{\\frac{1}{N}} \\sum_{i=1}^N \\Big( (\\mathbf{1}[y_i = +1] - 1)\\mathbf{w}^T h(\\mathbf{x}_i) - \\ln\\left(1 + \\exp(-\\mathbf{w}^T h(\\mathbf{x}_i))\\right) \\Big) $$\n",
    "\n",
    "**Note** that we made one tiny modification to the log likelihood function (called **compute_log_likelihood**) in our earlier assignments. We added a $\\color{red}{1/N}$ term which averages the log likelihood accross all data points. The $\\color{red}{1/N}$ term makes it easier for us to compare stochastic gradient ascent with batch gradient ascent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def compute_avg_log_likelihood(feature_matrix, sentiment, coefficients):\n",
    "    \n",
    "    indicator = (sentiment==+1)\n",
    "    scores = np.dot(feature_matrix, coefficients)\n",
    "    logexp = np.log(1. + np.exp(-scores))\n",
    "    \n",
    "    # Simple check to prevent overflow\n",
    "    mask = np.isinf(logexp)\n",
    "    logexp[mask] = -scores[mask]\n",
    "    \n",
    "    lp = np.sum((indicator-1)*scores - logexp)/len(feature_matrix)\n",
    "    \n",
    "    return lp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modifying the derivative for stochastic gradient ascent\n",
    "\n",
    "Recall that the gradient for a single data point $\\color{red}{\\mathbf{x}_i}$ can be computed using the following formula:\n",
    "\n",
    "$$\n",
    "\\frac{\\partial\\ell_{\\color{red}{i}}(\\mathbf{w})}{\\partial w_j} = h_j(\\color{red}{\\mathbf{x}_i})\\left(\\mathbf{1}[y_\\color{red}{i} = +1] - P(y_\\color{red}{i} = +1 | \\color{red}{\\mathbf{x}_i}, \\mathbf{w})\\right)\n",
    "$$\n",
    "\n",
    "\n",
    "** Computing the gradient for a single data point**\n",
    "\n",
    "Using NumPy, we access $\\mathbf{x}_i$ in the training data using `feature_matrix_train[i:i+1,:]`\n",
    "and $y_i$ in the training data using `sentiment_train[i:i+1]`. We can compute $\\partial\\ell_{\\color{red}{i}}(\\mathbf{w})/\\partial w_j$ by re-using **all the code** written in **feature_derivative** and **predict_probability**.\n",
    "\n",
    "\n",
    "We compute $\\partial\\ell_{\\color{red}{i}}(\\mathbf{w})/\\partial w_j$ using the following steps:\n",
    "* First, compute $P(y_i = +1 | \\mathbf{x}_i, \\mathbf{w})$ using the **predict_probability** function with `feature_matrix_train[i:i+1,:]` as the first parameter.\n",
    "* Next, compute $\\mathbf{1}[y_i = +1]$ using `sentiment_train[i:i+1]`.\n",
    "* Finally, call the **feature_derivative** function with `feature_matrix_train[i:i+1, j]` as one of the parameters. \n",
    "\n",
    "Let us follow these steps for `j = 1` and `i = 10`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gradient single data point: 0.0\n"
     ]
    }
   ],
   "source": [
    "j = 1                        # Feature number\n",
    "i = 10                       # Data point number\n",
    "coefficients = np.zeros(194) # A point w at which we are computing the gradient.\n",
    "\n",
    "predictions = predict_probability(feature_matrix_train[i:i+1,:], coefficients)\n",
    "indicator = (sentiment_train[i:i+1]==+1)\n",
    "\n",
    "errors = indicator - predictions        \n",
    "gradient_single_data_point = feature_derivative(errors, feature_matrix_train[i:i+1,j])\n",
    "print \"Gradient single data point: %s\" % gradient_single_data_point"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modifying the derivative for using a batch of data points\n",
    "\n",
    "Stochastic gradient estimates the ascent direction using 1 data point, while gradient uses $N$ data points to decide how to update the parameters. One simple change allows us to use a **mini-batch** of $B \\leq N$ data points to estimate the ascent direction. This simple approach is faster than regular gradient but less noisy than stochastic gradient that uses only 1 data point. \n",
    "\n",
    "Given a mini-batch (or a set of data points) $\\mathbf{x}_{i}, \\mathbf{x}_{i+1} \\ldots \\mathbf{x}_{i+B}$, the gradient function for this mini-batch of data points is given by:\n",
    "$$\n",
    "\\color{red}{\\sum_{s = i}^{i+B}} \\frac{\\partial\\ell_{s}}{\\partial w_j} = \\color{red}{\\sum_{s = i}^{i + B}} h_j(\\mathbf{x}_s)\\left(\\mathbf{1}[y_s = +1] - P(y_s = +1 | \\mathbf{x}_s, \\mathbf{w})\\right)\n",
    "$$\n",
    "\n",
    "\n",
    "** Computing the gradient for a \"mini-batch\" of data points**\n",
    "\n",
    "Using NumPy, we access the points $\\mathbf{x}_i, \\mathbf{x}_{i+1} \\ldots \\mathbf{x}_{i+B}$ in the training data using `feature_matrix_train[i:i+B,:]`\n",
    "and $y_i$ in the training data using `sentiment_train[i:i+B]`. \n",
    "\n",
    "We can compute $\\color{red}{\\sum_{s = i}^{i+B}} \\partial\\ell_{s}/\\partial w_j$ easily as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gradient mini-batch data points: 1.0\n"
     ]
    }
   ],
   "source": [
    "j = 1                        # Feature number\n",
    "i = 10                       # Data point start\n",
    "B = 10                       # Mini-batch size\n",
    "coefficients = np.zeros(194) # A point w at which we are computing the gradient.\n",
    "\n",
    "predictions = predict_probability(feature_matrix_train[i:i+B,:], coefficients)\n",
    "indicator = (sentiment_train[i:i+B]==+1)\n",
    "\n",
    "errors = indicator - predictions        \n",
    "gradient_mini_batch = feature_derivative(errors, feature_matrix_train[i:i+B,j])\n",
    "print \"Gradient mini-batch data points: %s\" % gradient_mini_batch\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Averaging the gradient across a batch\n",
    "\n",
    "It is a common practice to normalize the gradient update rule by the batch size B:\n",
    "\n",
    "$$\n",
    "\\frac{\\partial\\ell_{\\color{red}{A}}(\\mathbf{w})}{\\partial w_j} \\approx \\color{red}{\\frac{1}{B}} {\\sum_{s = i}^{i + B}} h_j(\\mathbf{x}_s)\\left(\\mathbf{1}[y_s = +1] - P(y_s = +1 | \\mathbf{x}_s, \\mathbf{w})\\right)\n",
    "$$\n",
    "In other words, we update the coefficients using the **average gradient over data points** (instead of using a summation). By using the average gradient, we ensure that the magnitude of the gradient is approximately the same for all batch sizes. This way, we can more easily compare various batch sizes of stochastic gradient ascent (including a batch size of **all the data points**), and study the effect of batch size on the algorithm as well as the choice of step size.\n",
    "\n",
    "\n",
    "## Implementing stochastic gradient ascent\n",
    "\n",
    "Now we are ready to implement our own logistic regression with stochastic gradient ascent. The following function fits a logistic regression model using gradient ascent:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from math import sqrt\n",
    "def logistic_regression_SG(feature_matrix, sentiment, initial_coefficients, step_size, batch_size, max_iter):\n",
    "    log_likelihood_all = []\n",
    "    \n",
    "    # make sure it's a numpy array\n",
    "    coefficients = np.array(initial_coefficients)\n",
    "    # set seed=1 to produce consistent results\n",
    "    np.random.seed(seed=1)\n",
    "    # Shuffle the data before starting\n",
    "    permutation = np.random.permutation(len(feature_matrix))\n",
    "    feature_matrix = feature_matrix[permutation,:]\n",
    "    sentiment = sentiment[permutation]\n",
    "    \n",
    "    i = 0 # index of current batch\n",
    "    # Do a linear scan over data\n",
    "    for itr in xrange(max_iter):\n",
    "        # Predict P(y_i = +1|x_i,w) using your predict_probability() function\n",
    "        # Make sure to slice the i-th row of feature_matrix with [i:i+batch_size,:]\n",
    "        predictions = predict_probability(feature_matrix[i:i+batch_size,:], coefficients)\n",
    "        \n",
    "        # Compute indicator value for (y_i = +1)\n",
    "        # Make sure to slice the i-th entry with [i:i+batch_size]\n",
    "        indicator = (sentiment[i:i+batch_size]==1)\n",
    "        \n",
    "        # Compute the errors as indicator - predictions\n",
    "        errors = indicator - predictions\n",
    "        for j in xrange(len(coefficients)): # loop over each coefficient\n",
    "            # Recall that feature_matrix[:,j] is the feature column associated with coefficients[j]\n",
    "            # Compute the derivative for coefficients[j] and save it to derivative.\n",
    "            # Make sure to slice the i-th row of feature_matrix with [i:i+batch_size,j]\n",
    "            derivative = feature_derivative(feature_matrix[i:i+batch_size,j], errors)\n",
    "            \n",
    "            # compute the product of the step size, the derivative, and the **normalization constant** (1./batch_size)\n",
    "            coefficients[j] += step_size * derivative * (1./batch_size)\n",
    "        \n",
    "        # Checking whether log likelihood is increasing\n",
    "        # Print the log likelihood over the *current batch*\n",
    "        lp = compute_avg_log_likelihood(feature_matrix[i:i+batch_size,:], sentiment[i:i+batch_size],\n",
    "                                        coefficients)\n",
    "        log_likelihood_all.append(lp)\n",
    "        if itr <= 15 or (itr <= 1000 and itr % 100 == 0) or (itr <= 10000 and itr % 1000 == 0) \\\n",
    "         or itr % 10000 == 0 or itr == max_iter-1:\n",
    "            data_size = len(feature_matrix)\n",
    "            print 'Iteration %*d: Average log likelihood (of data points in batch [%0*d:%0*d]) = %.8f' % \\\n",
    "                (int(np.ceil(np.log10(max_iter))), itr, \\\n",
    "                 int(np.ceil(np.log10(data_size))), i, \\\n",
    "                 int(np.ceil(np.log10(data_size))), i+batch_size, lp)\n",
    "        \n",
    "        # if we made a complete pass over data, shuffle and restart\n",
    "        i += batch_size\n",
    "        if i+batch_size > len(feature_matrix):\n",
    "            permutation = np.random.permutation(len(feature_matrix))\n",
    "            feature_matrix = feature_matrix[permutation,:]\n",
    "            sentiment = sentiment[permutation]\n",
    "            i = 0\n",
    "                \n",
    "    # We return the list of log likelihoods for plotting purposes.\n",
    "    return coefficients, log_likelihood_all"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Note**. In practice, the final set of coefficients is rarely used; it is better to use the average of the last K sets of coefficients instead, where K should be adjusted depending on how fast the log likelihood oscillates around the optimum."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Running gradient ascent using the stochastic gradient ascent implementation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Instead of implementing batch gradient ascent separately, we save time by re-using the stochastic gradient ascent function we just wrote &mdash; **to perform gradient ascent**, it suffices to set **`batch_size`** to the number of data points in the training data. \n",
    "\n",
    "**Small Caveat**. The batch gradient ascent implementation here is slightly different than the one in the earlier assignments, as we now normalize the gradient update rule.\n",
    "\n",
    "We now **run stochastic gradient ascent** over the **feature_matrix_train** for 10 iterations using:\n",
    "* `initial_coefficients = np.zeros(194)`\n",
    "* `step_size = 5e-1`\n",
    "* `batch_size = 1`\n",
    "* `max_iter = 10`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration 0: Average log likelihood (of data points in batch [00000:00001]) = -0.25192908\n",
      "Iteration 1: Average log likelihood (of data points in batch [00001:00002]) = -0.00000001\n",
      "Iteration 2: Average log likelihood (of data points in batch [00002:00003]) = -0.12692771\n",
      "Iteration 3: Average log likelihood (of data points in batch [00003:00004]) = -0.02969101\n",
      "Iteration 4: Average log likelihood (of data points in batch [00004:00005]) = -0.02668819\n",
      "Iteration 5: Average log likelihood (of data points in batch [00005:00006]) = -0.04332901\n",
      "Iteration 6: Average log likelihood (of data points in batch [00006:00007]) = -0.02368802\n",
      "Iteration 7: Average log likelihood (of data points in batch [00007:00008]) = -0.12686897\n",
      "Iteration 8: Average log likelihood (of data points in batch [00008:00009]) = -0.04468879\n",
      "Iteration 9: Average log likelihood (of data points in batch [00009:00010]) = -0.00000124\n"
     ]
    }
   ],
   "source": [
    "coefficients, log_likelihood = logistic_regression_SG(feature_matrix_train, sentiment_train,\n",
    "                                        initial_coefficients=np.zeros(194),\n",
    "                                        step_size=5e-1, batch_size=1, max_iter=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now run **batch gradient ascent** over the **feature_matrix_train** for 200 iterations using:\n",
    "* `initial_coefficients = np.zeros(194)`\n",
    "* `step_size = 5e-1`\n",
    "* `batch_size = len(feature_matrix_train)`\n",
    "* `max_iter = 200`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration   0: Average log likelihood (of data points in batch [00000:47780]) = -0.68308119\n",
      "Iteration   1: Average log likelihood (of data points in batch [00000:47780]) = -0.67394599\n",
      "Iteration   2: Average log likelihood (of data points in batch [00000:47780]) = -0.66555129\n",
      "Iteration   3: Average log likelihood (of data points in batch [00000:47780]) = -0.65779626\n",
      "Iteration   4: Average log likelihood (of data points in batch [00000:47780]) = -0.65060701\n",
      "Iteration   5: Average log likelihood (of data points in batch [00000:47780]) = -0.64392241\n",
      "Iteration   6: Average log likelihood (of data points in batch [00000:47780]) = -0.63769009\n",
      "Iteration   7: Average log likelihood (of data points in batch [00000:47780]) = -0.63186462\n",
      "Iteration   8: Average log likelihood (of data points in batch [00000:47780]) = -0.62640636\n",
      "Iteration   9: Average log likelihood (of data points in batch [00000:47780]) = -0.62128063\n",
      "Iteration  10: Average log likelihood (of data points in batch [00000:47780]) = -0.61645691\n",
      "Iteration  11: Average log likelihood (of data points in batch [00000:47780]) = -0.61190832\n",
      "Iteration  12: Average log likelihood (of data points in batch [00000:47780]) = -0.60761103\n",
      "Iteration  13: Average log likelihood (of data points in batch [00000:47780]) = -0.60354390\n",
      "Iteration  14: Average log likelihood (of data points in batch [00000:47780]) = -0.59968811\n",
      "Iteration  15: Average log likelihood (of data points in batch [00000:47780]) = -0.59602682\n",
      "Iteration 100: Average log likelihood (of data points in batch [00000:47780]) = -0.49520194\n",
      "Iteration 199: Average log likelihood (of data points in batch [00000:47780]) = -0.47126953\n"
     ]
    }
   ],
   "source": [
    "coefficients_batch, log_likelihood_batch = logistic_regression_SG(feature_matrix_train, sentiment_train,\n",
    "                                        initial_coefficients=np.zeros(194),\n",
    "                                        step_size=5e-1, batch_size=len(feature_matrix_train), max_iter=200)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Make \"passes\" over the dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To make a fair comparison betweeen stochastic gradient ascent and batch gradient ascent, we measure the average log likelihood as a function of the number of passes (defined as follows):\n",
    "$$\n",
    "[\\text{# of passes}] = \\frac{[\\text{# of data points touched so far}]}{[\\text{size of dataset}]}\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Suppose that we run stochastic gradient ascent with a batch size of 100. How many gradient updates are performed at the end of two passes over a dataset consisting of 50000 data points?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1000.0"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "2*(50000/100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Log likelihood plots for stochastic gradient ascent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With the terminology in mind, let us run stochastic gradient ascent for 10 passes. We will use\n",
    "* `step_size=1e-1`\n",
    "* `batch_size=100`\n",
    "* `initial_coefficients` to all zeros."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration    0: Average log likelihood (of data points in batch [00000:00100]) = -0.68251093\n",
      "Iteration    1: Average log likelihood (of data points in batch [00100:00200]) = -0.67845294\n",
      "Iteration    2: Average log likelihood (of data points in batch [00200:00300]) = -0.68207160\n",
      "Iteration    3: Average log likelihood (of data points in batch [00300:00400]) = -0.67411325\n",
      "Iteration    4: Average log likelihood (of data points in batch [00400:00500]) = -0.67804438\n",
      "Iteration    5: Average log likelihood (of data points in batch [00500:00600]) = -0.67712546\n",
      "Iteration    6: Average log likelihood (of data points in batch [00600:00700]) = -0.66377074\n",
      "Iteration    7: Average log likelihood (of data points in batch [00700:00800]) = -0.67321231\n",
      "Iteration    8: Average log likelihood (of data points in batch [00800:00900]) = -0.66923613\n",
      "Iteration    9: Average log likelihood (of data points in batch [00900:01000]) = -0.67479446\n",
      "Iteration   10: Average log likelihood (of data points in batch [01000:01100]) = -0.66501639\n",
      "Iteration   11: Average log likelihood (of data points in batch [01100:01200]) = -0.65591964\n",
      "Iteration   12: Average log likelihood (of data points in batch [01200:01300]) = -0.66240398\n",
      "Iteration   13: Average log likelihood (of data points in batch [01300:01400]) = -0.66440641\n",
      "Iteration   14: Average log likelihood (of data points in batch [01400:01500]) = -0.65782757\n",
      "Iteration   15: Average log likelihood (of data points in batch [01500:01600]) = -0.64571479\n",
      "Iteration  100: Average log likelihood (of data points in batch [10000:10100]) = -0.60976663\n",
      "Iteration  200: Average log likelihood (of data points in batch [20000:20100]) = -0.54566060\n",
      "Iteration  300: Average log likelihood (of data points in batch [30000:30100]) = -0.48245740\n",
      "Iteration  400: Average log likelihood (of data points in batch [40000:40100]) = -0.46629313\n",
      "Iteration  500: Average log likelihood (of data points in batch [02300:02400]) = -0.47223389\n",
      "Iteration  600: Average log likelihood (of data points in batch [12300:12400]) = -0.52216798\n",
      "Iteration  700: Average log likelihood (of data points in batch [22300:22400]) = -0.52336683\n",
      "Iteration  800: Average log likelihood (of data points in batch [32300:32400]) = -0.46963453\n",
      "Iteration  900: Average log likelihood (of data points in batch [42300:42400]) = -0.47883783\n",
      "Iteration 1000: Average log likelihood (of data points in batch [04600:04700]) = -0.46988191\n",
      "Iteration 2000: Average log likelihood (of data points in batch [09200:09300]) = -0.46365531\n",
      "Iteration 3000: Average log likelihood (of data points in batch [13800:13900]) = -0.36466901\n",
      "Iteration 4000: Average log likelihood (of data points in batch [18400:18500]) = -0.51096892\n",
      "Iteration 4769: Average log likelihood (of data points in batch [47600:47700]) = -0.54670667\n"
     ]
    }
   ],
   "source": [
    "step_size = 1e-1\n",
    "batch_size = 100\n",
    "num_passes = 10\n",
    "num_iterations = num_passes * int(len(feature_matrix_train)/batch_size)\n",
    "\n",
    "coefficients_sgd, log_likelihood_sgd = logistic_regression_SG(feature_matrix_train, sentiment_train,\n",
    "                                       initial_coefficients=np.zeros(194),\n",
    "                                       step_size=1e-1, batch_size=100, max_iter=num_iterations)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We run the followiing utility function to plot the average log likelihood as a function of the number of passes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "def make_plot(log_likelihood_all, len_data, batch_size, smoothing_window=1, label=''):\n",
    "    plt.rcParams.update({'figure.figsize': (9,5)})\n",
    "    log_likelihood_all_ma = np.convolve(np.array(log_likelihood_all), \\\n",
    "                                        np.ones((smoothing_window,))/smoothing_window, mode='valid')\n",
    "    plt.plot(np.array(range(smoothing_window-1, len(log_likelihood_all)))*float(batch_size)/len_data,\n",
    "             log_likelihood_all_ma, linewidth=4.0, label=label)\n",
    "    plt.rcParams.update({'font.size': 16})\n",
    "    plt.tight_layout()\n",
    "    plt.xlabel('# of passes over data')\n",
    "    plt.ylabel('Average log likelihood per data point')\n",
    "    plt.legend(loc='lower right', prop={'size':14})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4MAAAHxCAYAAADX4mkZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x117d13750>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "make_plot(log_likelihood_sgd, len_data=len(feature_matrix_train), batch_size=100,\n",
    "          label='stochastic gradient, step_size=1e-1')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Smoothing the stochastic gradient ascent curve\n",
    "\n",
    "The plotted line oscillates so much that it is hard to see whether the log likelihood is improving. In our plot, we apply a simple smoothing operation using the parameter `smoothing_window`. The smoothing is simply a [moving average](https://en.wikipedia.org/wiki/Moving_average) of log likelihood over the last `smoothing_window` \"iterations\" of  stochastic gradient ascent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4MAAAHxCAYAAADX4mkZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x122ceb790>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "make_plot(log_likelihood_sgd, len_data=len(feature_matrix_train), batch_size=100,\n",
    "          smoothing_window=30, label='stochastic gradient, step_size=1e-1')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Stochastic gradient ascent vs batch gradient ascent\n",
    "\n",
    "To compare convergence rates for stochastic gradient ascent with batch gradient ascent, we call `make_plot()` multiple times in the same cell.\n",
    "\n",
    "We are comparing:\n",
    "* **stochastic gradient ascent**: `step_size = 0.1`, `batch_size=100`\n",
    "* **batch gradient ascent**: `step_size = 0.5`, `batch_size=len(feature_matrix_train)`\n",
    "\n",
    "Write code to run stochastic gradient ascent for 200 passes using:\n",
    "* `step_size=1e-1`\n",
    "* `batch_size=100`\n",
    "* `initial_coefficients` to all zeros."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration     0: Average log likelihood (of data points in batch [00000:00100]) = -0.68251093\n",
      "Iteration     1: Average log likelihood (of data points in batch [00100:00200]) = -0.67845294\n",
      "Iteration     2: Average log likelihood (of data points in batch [00200:00300]) = -0.68207160\n",
      "Iteration     3: Average log likelihood (of data points in batch [00300:00400]) = -0.67411325\n",
      "Iteration     4: Average log likelihood (of data points in batch [00400:00500]) = -0.67804438\n",
      "Iteration     5: Average log likelihood (of data points in batch [00500:00600]) = -0.67712546\n",
      "Iteration     6: Average log likelihood (of data points in batch [00600:00700]) = -0.66377074\n",
      "Iteration     7: Average log likelihood (of data points in batch [00700:00800]) = -0.67321231\n",
      "Iteration     8: Average log likelihood (of data points in batch [00800:00900]) = -0.66923613\n",
      "Iteration     9: Average log likelihood (of data points in batch [00900:01000]) = -0.67479446\n",
      "Iteration    10: Average log likelihood (of data points in batch [01000:01100]) = -0.66501639\n",
      "Iteration    11: Average log likelihood (of data points in batch [01100:01200]) = -0.65591964\n",
      "Iteration    12: Average log likelihood (of data points in batch [01200:01300]) = -0.66240398\n",
      "Iteration    13: Average log likelihood (of data points in batch [01300:01400]) = -0.66440641\n",
      "Iteration    14: Average log likelihood (of data points in batch [01400:01500]) = -0.65782757\n",
      "Iteration    15: Average log likelihood (of data points in batch [01500:01600]) = -0.64571479\n",
      "Iteration   100: Average log likelihood (of data points in batch [10000:10100]) = -0.60976663\n",
      "Iteration   200: Average log likelihood (of data points in batch [20000:20100]) = -0.54566060\n",
      "Iteration   300: Average log likelihood (of data points in batch [30000:30100]) = -0.48245740\n",
      "Iteration   400: Average log likelihood (of data points in batch [40000:40100]) = -0.46629313\n",
      "Iteration   500: Average log likelihood (of data points in batch [02300:02400]) = -0.47223389\n",
      "Iteration   600: Average log likelihood (of data points in batch [12300:12400]) = -0.52216798\n",
      "Iteration   700: Average log likelihood (of data points in batch [22300:22400]) = -0.52336683\n",
      "Iteration   800: Average log likelihood (of data points in batch [32300:32400]) = -0.46963453\n",
      "Iteration   900: Average log likelihood (of data points in batch [42300:42400]) = -0.47883783\n",
      "Iteration  1000: Average log likelihood (of data points in batch [04600:04700]) = -0.46988191\n",
      "Iteration  2000: Average log likelihood (of data points in batch [09200:09300]) = -0.46365531\n",
      "Iteration  3000: Average log likelihood (of data points in batch [13800:13900]) = -0.36466901\n",
      "Iteration  4000: Average log likelihood (of data points in batch [18400:18500]) = -0.51096892\n",
      "Iteration  5000: Average log likelihood (of data points in batch [23000:23100]) = -0.43544394\n",
      "Iteration  6000: Average log likelihood (of data points in batch [27600:27700]) = -0.45656653\n",
      "Iteration  7000: Average log likelihood (of data points in batch [32200:32300]) = -0.42656766\n",
      "Iteration  8000: Average log likelihood (of data points in batch [36800:36900]) = -0.39989352\n",
      "Iteration  9000: Average log likelihood (of data points in batch [41400:41500]) = -0.45267388\n",
      "Iteration 10000: Average log likelihood (of data points in batch [46000:46100]) = -0.45394262\n",
      "Iteration 20000: Average log likelihood (of data points in batch [44300:44400]) = -0.48958438\n",
      "Iteration 30000: Average log likelihood (of data points in batch [42600:42700]) = -0.41913672\n",
      "Iteration 40000: Average log likelihood (of data points in batch [40900:41000]) = -0.45899229\n",
      "Iteration 50000: Average log likelihood (of data points in batch [39200:39300]) = -0.46859254\n",
      "Iteration 60000: Average log likelihood (of data points in batch [37500:37600]) = -0.41599369\n",
      "Iteration 70000: Average log likelihood (of data points in batch [35800:35900]) = -0.49905981\n",
      "Iteration 80000: Average log likelihood (of data points in batch [34100:34200]) = -0.45494095\n",
      "Iteration 90000: Average log likelihood (of data points in batch [32400:32500]) = -0.43220080\n",
      "Iteration 95399: Average log likelihood (of data points in batch [47600:47700]) = -0.50265709\n"
     ]
    }
   ],
   "source": [
    "step_size = 1e-1\n",
    "batch_size = 100\n",
    "num_passes = 200\n",
    "num_iterations = num_passes * int(len(feature_matrix_train)/batch_size)\n",
    "\n",
    "coefficients_sgd, log_likelihood_sgd = logistic_regression_SG(feature_matrix_train, sentiment_train,\n",
    "                                       initial_coefficients=np.zeros(194),\n",
    "                                       step_size=1e-1, batch_size=100, max_iter=num_iterations)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We compare the convergence of stochastic gradient ascent and batch gradient ascent in the following cell. Note that we apply smoothing with `smoothing_window=30`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2cAAAHYCAYAAADJSJDsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x122d4e3d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "make_plot(log_likelihood_sgd, len_data=len(feature_matrix_train), batch_size=100,\n",
    "          smoothing_window=30, label='stochastic, step_size=1e-1')\n",
    "make_plot(log_likelihood_batch, len_data=len(feature_matrix_train), batch_size=len(feature_matrix_train),\n",
    "          smoothing_window=1, label='batch, step_size=5e-1')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Explore the effects of step sizes on stochastic gradient ascent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To start, we explore a wide range of step sizes that are equally spaced in the log space. Run stochastic gradient ascent with `step_size` set to 1e-4, 1e-3, 1e-2, 1e-1, 1e0, 1e1, and 1e2. Use the following set of parameters:\n",
    "* `initial_coefficients=np.zeros(194)`\n",
    "* `batch_size=100`\n",
    "* `max_iter` initialized so as to run 10 passes over the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration    0: Average log likelihood (of data points in batch [00000:00100]) = -0.69313622\n",
      "Iteration    1: Average log likelihood (of data points in batch [00100:00200]) = -0.69313170\n",
      "Iteration    2: Average log likelihood (of data points in batch [00200:00300]) = -0.69313585\n",
      "Iteration    3: Average log likelihood (of data points in batch [00300:00400]) = -0.69312487\n",
      "Iteration    4: Average log likelihood (of data points in batch [00400:00500]) = -0.69313157\n",
      "Iteration    5: Average log likelihood (of data points in batch [00500:00600]) = -0.69313113\n",
      "Iteration    6: Average log likelihood (of data points in batch [00600:00700]) = -0.69311121\n",
      "Iteration    7: Average log likelihood (of data points in batch [00700:00800]) = -0.69312692\n",
      "Iteration    8: Average log likelihood (of data points in batch [00800:00900]) = -0.69312115\n",
      "Iteration    9: Average log likelihood (of data points in batch [00900:01000]) = -0.69312811\n",
      "Iteration   10: Average log likelihood (of data points in batch [01000:01100]) = -0.69311286\n",
      "Iteration   11: Average log likelihood (of data points in batch [01100:01200]) = -0.69310301\n",
      "Iteration   12: Average log likelihood (of data points in batch [01200:01300]) = -0.69310725\n",
      "Iteration   13: Average log likelihood (of data points in batch [01300:01400]) = -0.69311567\n",
      "Iteration   14: Average log likelihood (of data points in batch [01400:01500]) = -0.69310836\n",
      "Iteration   15: Average log likelihood (of data points in batch [01500:01600]) = -0.69308342\n",
      "Iteration  100: Average log likelihood (of data points in batch [10000:10100]) = -0.69298918\n",
      "Iteration  200: Average log likelihood (of data points in batch [20000:20100]) = -0.69277472\n",
      "Iteration  300: Average log likelihood (of data points in batch [30000:30100]) = -0.69228764\n",
      "Iteration  400: Average log likelihood (of data points in batch [40000:40100]) = -0.69222554\n",
      "Iteration  500: Average log likelihood (of data points in batch [02300:02400]) = -0.69186710\n",
      "Iteration  600: Average log likelihood (of data points in batch [12300:12400]) = -0.69230650\n",
      "Iteration  700: Average log likelihood (of data points in batch [22300:22400]) = -0.69174220\n",
      "Iteration  800: Average log likelihood (of data points in batch [32300:32400]) = -0.69139955\n",
      "Iteration  900: Average log likelihood (of data points in batch [42300:42400]) = -0.69123818\n",
      "Iteration 1000: Average log likelihood (of data points in batch [04600:04700]) = -0.69088883\n",
      "Iteration 2000: Average log likelihood (of data points in batch [09200:09300]) = -0.68976850\n",
      "Iteration 3000: Average log likelihood (of data points in batch [13800:13900]) = -0.68569701\n",
      "Iteration 4000: Average log likelihood (of data points in batch [18400:18500]) = -0.68597545\n",
      "Iteration 4769: Average log likelihood (of data points in batch [47600:47700]) = -0.68736824\n",
      "Iteration    0: Average log likelihood (of data points in batch [00000:00100]) = -0.69303759\n",
      "Iteration    1: Average log likelihood (of data points in batch [00100:00200]) = -0.69299241\n",
      "Iteration    2: Average log likelihood (of data points in batch [00200:00300]) = -0.69303389\n",
      "Iteration    3: Average log likelihood (of data points in batch [00300:00400]) = -0.69292442\n",
      "Iteration    4: Average log likelihood (of data points in batch [00400:00500]) = -0.69299113\n",
      "Iteration    5: Average log likelihood (of data points in batch [00500:00600]) = -0.69298668\n",
      "Iteration    6: Average log likelihood (of data points in batch [00600:00700]) = -0.69278828\n",
      "Iteration    7: Average log likelihood (of data points in batch [00700:00800]) = -0.69294460\n",
      "Iteration    8: Average log likelihood (of data points in batch [00800:00900]) = -0.69288708\n",
      "Iteration    9: Average log likelihood (of data points in batch [00900:01000]) = -0.69295651\n",
      "Iteration   10: Average log likelihood (of data points in batch [01000:01100]) = -0.69280480\n",
      "Iteration   11: Average log likelihood (of data points in batch [01100:01200]) = -0.69270635\n",
      "Iteration   12: Average log likelihood (of data points in batch [01200:01300]) = -0.69274924\n",
      "Iteration   13: Average log likelihood (of data points in batch [01300:01400]) = -0.69283249\n",
      "Iteration   14: Average log likelihood (of data points in batch [01400:01500]) = -0.69275924\n",
      "Iteration   15: Average log likelihood (of data points in batch [01500:01600]) = -0.69251197\n",
      "Iteration  100: Average log likelihood (of data points in batch [10000:10100]) = -0.69158805\n",
      "Iteration  200: Average log likelihood (of data points in batch [20000:20100]) = -0.68946852\n",
      "Iteration  300: Average log likelihood (of data points in batch [30000:30100]) = -0.68492418\n",
      "Iteration  400: Average log likelihood (of data points in batch [40000:40100]) = -0.68415366\n",
      "Iteration  500: Average log likelihood (of data points in batch [02300:02400]) = -0.68114554\n",
      "Iteration  600: Average log likelihood (of data points in batch [12300:12400]) = -0.68489867\n",
      "Iteration  700: Average log likelihood (of data points in batch [22300:22400]) = -0.68027821\n",
      "Iteration  800: Average log likelihood (of data points in batch [32300:32400]) = -0.67693088\n",
      "Iteration  900: Average log likelihood (of data points in batch [42300:42400]) = -0.67561867\n",
      "Iteration 1000: Average log likelihood (of data points in batch [04600:04700]) = -0.67367588\n",
      "Iteration 2000: Average log likelihood (of data points in batch [09200:09300]) = -0.66156206\n",
      "Iteration 3000: Average log likelihood (of data points in batch [13800:13900]) = -0.62798175\n",
      "Iteration 4000: Average log likelihood (of data points in batch [18400:18500]) = -0.64157978\n",
      "Iteration 4769: Average log likelihood (of data points in batch [47600:47700]) = -0.64571292\n",
      "Iteration    0: Average log likelihood (of data points in batch [00000:00100]) = -0.69205420\n",
      "Iteration    1: Average log likelihood (of data points in batch [00100:00200]) = -0.69160695\n",
      "Iteration    2: Average log likelihood (of data points in batch [00200:00300]) = -0.69201686\n",
      "Iteration    3: Average log likelihood (of data points in batch [00300:00400]) = -0.69095428\n",
      "Iteration    4: Average log likelihood (of data points in batch [00400:00500]) = -0.69159348\n",
      "Iteration    5: Average log likelihood (of data points in batch [00500:00600]) = -0.69154386\n",
      "Iteration    6: Average log likelihood (of data points in batch [00600:00700]) = -0.68964000\n",
      "Iteration    7: Average log likelihood (of data points in batch [00700:00800]) = -0.69112685\n",
      "Iteration    8: Average log likelihood (of data points in batch [00800:00900]) = -0.69056997\n",
      "Iteration    9: Average log likelihood (of data points in batch [00900:01000]) = -0.69124730\n",
      "Iteration   10: Average log likelihood (of data points in batch [01000:01100]) = -0.68980179\n",
      "Iteration   11: Average log likelihood (of data points in batch [01100:01200]) = -0.68882576\n",
      "Iteration   12: Average log likelihood (of data points in batch [01200:01300]) = -0.68929536\n",
      "Iteration   13: Average log likelihood (of data points in batch [01300:01400]) = -0.69003572\n",
      "Iteration   14: Average log likelihood (of data points in batch [01400:01500]) = -0.68929307\n",
      "Iteration   15: Average log likelihood (of data points in batch [01500:01600]) = -0.68702353\n",
      "Iteration  100: Average log likelihood (of data points in batch [10000:10100]) = -0.67916061\n",
      "Iteration  200: Average log likelihood (of data points in batch [20000:20100]) = -0.66049079\n",
      "Iteration  300: Average log likelihood (of data points in batch [30000:30100]) = -0.63235099\n",
      "Iteration  400: Average log likelihood (of data points in batch [40000:40100]) = -0.62183600\n",
      "Iteration  500: Average log likelihood (of data points in batch [02300:02400]) = -0.61150928\n",
      "Iteration  600: Average log likelihood (of data points in batch [12300:12400]) = -0.62979300\n",
      "Iteration  700: Average log likelihood (of data points in batch [22300:22400]) = -0.61553432\n",
      "Iteration  800: Average log likelihood (of data points in batch [32300:32400]) = -0.59156014\n",
      "Iteration  900: Average log likelihood (of data points in batch [42300:42400]) = -0.58842264\n",
      "Iteration 1000: Average log likelihood (of data points in batch [04600:04700]) = -0.59076267\n",
      "Iteration 2000: Average log likelihood (of data points in batch [09200:09300]) = -0.54480104\n",
      "Iteration 3000: Average log likelihood (of data points in batch [13800:13900]) = -0.45761063\n",
      "Iteration 4000: Average log likelihood (of data points in batch [18400:18500]) = -0.54362587\n",
      "Iteration 4769: Average log likelihood (of data points in batch [47600:47700]) = -0.56306510\n",
      "Iteration    0: Average log likelihood (of data points in batch [00000:00100]) = -0.68251093\n",
      "Iteration    1: Average log likelihood (of data points in batch [00100:00200]) = -0.67845294\n",
      "Iteration    2: Average log likelihood (of data points in batch [00200:00300]) = -0.68207160\n",
      "Iteration    3: Average log likelihood (of data points in batch [00300:00400]) = -0.67411325\n",
      "Iteration    4: Average log likelihood (of data points in batch [00400:00500]) = -0.67804438\n",
      "Iteration    5: Average log likelihood (of data points in batch [00500:00600]) = -0.67712546\n",
      "Iteration    6: Average log likelihood (of data points in batch [00600:00700]) = -0.66377074\n",
      "Iteration    7: Average log likelihood (of data points in batch [00700:00800]) = -0.67321231\n",
      "Iteration    8: Average log likelihood (of data points in batch [00800:00900]) = -0.66923613\n",
      "Iteration    9: Average log likelihood (of data points in batch [00900:01000]) = -0.67479446\n",
      "Iteration   10: Average log likelihood (of data points in batch [01000:01100]) = -0.66501639\n",
      "Iteration   11: Average log likelihood (of data points in batch [01100:01200]) = -0.65591964\n",
      "Iteration   12: Average log likelihood (of data points in batch [01200:01300]) = -0.66240398\n",
      "Iteration   13: Average log likelihood (of data points in batch [01300:01400]) = -0.66440641\n",
      "Iteration   14: Average log likelihood (of data points in batch [01400:01500]) = -0.65782757\n",
      "Iteration   15: Average log likelihood (of data points in batch [01500:01600]) = -0.64571479\n",
      "Iteration  100: Average log likelihood (of data points in batch [10000:10100]) = -0.60976663\n",
      "Iteration  200: Average log likelihood (of data points in batch [20000:20100]) = -0.54566060\n",
      "Iteration  300: Average log likelihood (of data points in batch [30000:30100]) = -0.48245740\n",
      "Iteration  400: Average log likelihood (of data points in batch [40000:40100]) = -0.46629313\n",
      "Iteration  500: Average log likelihood (of data points in batch [02300:02400]) = -0.47223389\n",
      "Iteration  600: Average log likelihood (of data points in batch [12300:12400]) = -0.52216798\n",
      "Iteration  700: Average log likelihood (of data points in batch [22300:22400]) = -0.52336683\n",
      "Iteration  800: Average log likelihood (of data points in batch [32300:32400]) = -0.46963453\n",
      "Iteration  900: Average log likelihood (of data points in batch [42300:42400]) = -0.47883783\n",
      "Iteration 1000: Average log likelihood (of data points in batch [04600:04700]) = -0.46988191\n",
      "Iteration 2000: Average log likelihood (of data points in batch [09200:09300]) = -0.46365531\n",
      "Iteration 3000: Average log likelihood (of data points in batch [13800:13900]) = -0.36466901\n",
      "Iteration 4000: Average log likelihood (of data points in batch [18400:18500]) = -0.51096892\n",
      "Iteration 4769: Average log likelihood (of data points in batch [47600:47700]) = -0.54670667\n",
      "Iteration    0: Average log likelihood (of data points in batch [00000:00100]) = -0.61201447\n",
      "Iteration    1: Average log likelihood (of data points in batch [00100:00200]) = -0.58843678\n",
      "Iteration    2: Average log likelihood (of data points in batch [00200:00300]) = -0.59771677\n",
      "Iteration    3: Average log likelihood (of data points in batch [00300:00400]) = -0.58770466\n",
      "Iteration    4: Average log likelihood (of data points in batch [00400:00500]) = -0.56939710\n",
      "Iteration    5: Average log likelihood (of data points in batch [00500:00600]) = -0.57554451\n",
      "Iteration    6: Average log likelihood (of data points in batch [00600:00700]) = -0.54068090\n",
      "Iteration    7: Average log likelihood (of data points in batch [00700:00800]) = -0.55212916\n",
      "Iteration    8: Average log likelihood (of data points in batch [00800:00900]) = -0.55311029\n",
      "Iteration    9: Average log likelihood (of data points in batch [00900:01000]) = -0.57672007\n",
      "Iteration   10: Average log likelihood (of data points in batch [01000:01100]) = -0.55455807\n",
      "Iteration   11: Average log likelihood (of data points in batch [01100:01200]) = -0.49771894\n",
      "Iteration   12: Average log likelihood (of data points in batch [01200:01300]) = -0.54708765\n",
      "Iteration   13: Average log likelihood (of data points in batch [01300:01400]) = -0.54286814\n",
      "Iteration   14: Average log likelihood (of data points in batch [01400:01500]) = -0.52361054\n",
      "Iteration   15: Average log likelihood (of data points in batch [01500:01600]) = -0.49731367\n",
      "Iteration  100: Average log likelihood (of data points in batch [10000:10100]) = -0.50102061\n",
      "Iteration  200: Average log likelihood (of data points in batch [20000:20100]) = -0.42406927\n",
      "Iteration  300: Average log likelihood (of data points in batch [30000:30100]) = -0.35064478\n",
      "Iteration  400: Average log likelihood (of data points in batch [40000:40100]) = -0.38344116\n",
      "Iteration  500: Average log likelihood (of data points in batch [02300:02400]) = -0.40170047\n",
      "Iteration  600: Average log likelihood (of data points in batch [12300:12400]) = -0.45117863\n",
      "Iteration  700: Average log likelihood (of data points in batch [22300:22400]) = -0.46493371\n",
      "Iteration  800: Average log likelihood (of data points in batch [32300:32400]) = -0.45343350\n",
      "Iteration  900: Average log likelihood (of data points in batch [42300:42400]) = -0.43128394\n",
      "Iteration 1000: Average log likelihood (of data points in batch [04600:04700]) = -0.43169967\n",
      "Iteration 2000: Average log likelihood (of data points in batch [09200:09300]) = -0.43029376\n",
      "Iteration 3000: Average log likelihood (of data points in batch [13800:13900]) = -0.32703099\n",
      "Iteration 4000: Average log likelihood (of data points in batch [18400:18500]) = -0.49162447\n",
      "Iteration 4769: Average log likelihood (of data points in batch [47600:47700]) = -0.52452720\n",
      "Iteration    0: Average log likelihood (of data points in batch [00000:00100]) = -0.51319004\n",
      "Iteration    1: Average log likelihood (of data points in batch [00100:00200]) = -2.20035379\n",
      "Iteration    2: Average log likelihood (of data points in batch [00200:00300]) = -3.34199720\n",
      "Iteration    3: Average log likelihood (of data points in batch [00300:00400]) = -3.06285156\n",
      "Iteration    4: Average log likelihood (of data points in batch [00400:00500]) = -2.80822162\n",
      "Iteration    5: Average log likelihood (of data points in batch [00500:00600]) = -2.99629286\n",
      "Iteration    6: Average log likelihood (of data points in batch [00600:00700]) = -2.71489944\n",
      "Iteration    7: Average log likelihood (of data points in batch [00700:00800]) = -3.61713200\n",
      "Iteration    8: Average log likelihood (of data points in batch [00800:00900]) = -1.19526584\n",
      "Iteration    9: Average log likelihood (of data points in batch [00900:01000]) = -0.75357081\n",
      "Iteration   10: Average log likelihood (of data points in batch [01000:01100]) = -0.71310829\n",
      "Iteration   11: Average log likelihood (of data points in batch [01100:01200]) = -0.59361318\n",
      "Iteration   12: Average log likelihood (of data points in batch [01200:01300]) = -1.53764659\n",
      "Iteration   13: Average log likelihood (of data points in batch [01300:01400]) = -2.69588686\n",
      "Iteration   14: Average log likelihood (of data points in batch [01400:01500]) = -1.89731473\n",
      "Iteration   15: Average log likelihood (of data points in batch [01500:01600]) = -0.81254441\n",
      "Iteration  100: Average log likelihood (of data points in batch [10000:10100]) = -1.19013437\n",
      "Iteration  200: Average log likelihood (of data points in batch [20000:20100]) = -0.48968363\n",
      "Iteration  300: Average log likelihood (of data points in batch [30000:30100]) = -0.72860037\n",
      "Iteration  400: Average log likelihood (of data points in batch [40000:40100]) = -0.58719556\n",
      "Iteration  500: Average log likelihood (of data points in batch [02300:02400]) = -0.31220572\n",
      "Iteration  600: Average log likelihood (of data points in batch [12300:12400]) = -1.89468446\n",
      "Iteration  700: Average log likelihood (of data points in batch [22300:22400]) = -0.96096585\n",
      "Iteration  800: Average log likelihood (of data points in batch [32300:32400]) = -0.66616640\n",
      "Iteration  900: Average log likelihood (of data points in batch [42300:42400]) = -0.46114004\n",
      "Iteration 1000: Average log likelihood (of data points in batch [04600:04700]) = -0.47236476\n",
      "Iteration 2000: Average log likelihood (of data points in batch [09200:09300]) = -0.45227508\n",
      "Iteration 3000: Average log likelihood (of data points in batch [13800:13900]) = -0.29378688\n",
      "Iteration 4000: Average log likelihood (of data points in batch [18400:18500]) = -2.47834692\n",
      "Iteration 4769: Average log likelihood (of data points in batch [47600:47700]) = -2.48776279\n",
      "Iteration    0: Average log likelihood (of data points in batch [00000:00100]) = -2.44471310\n",
      "Iteration    1: Average log likelihood (of data points in batch [00100:00200]) = -36.66862050\n",
      "Iteration    2: Average log likelihood (of data points in batch [00200:00300]) = -25.49870239\n",
      "Iteration    3: Average log likelihood (of data points in batch [00300:00400]) = -40.14565040\n",
      "Iteration    4: Average log likelihood (of data points in batch [00400:00500]) = -27.03748522\n",
      "Iteration    5: Average log likelihood (of data points in batch [00500:00600]) = -32.62294582\n",
      "Iteration    6: Average log likelihood (of data points in batch [00600:00700]) = -25.88017915\n",
      "Iteration    7: Average log likelihood (of data points in batch [00700:00800]) = -37.30720216\n",
      "Iteration    8: Average log likelihood (of data points in batch [00800:00900]) = -10.87360529\n",
      "Iteration    9: Average log likelihood (of data points in batch [00900:01000]) = -6.60878996\n",
      "Iteration   10: Average log likelihood (of data points in batch [01000:01100]) = -7.15375088\n",
      "Iteration   11: Average log likelihood (of data points in batch [01100:01200]) = -6.04741293\n",
      "Iteration   12: Average log likelihood (of data points in batch [01200:01300]) = -18.17389834\n",
      "Iteration   13: Average log likelihood (of data points in batch [01300:01400]) = -27.14619228\n",
      "Iteration   14: Average log likelihood (of data points in batch [01400:01500]) = -20.50685042\n",
      "Iteration   15: Average log likelihood (of data points in batch [01500:01600]) = -7.74332305\n",
      "Iteration  100: Average log likelihood (of data points in batch [10000:10100]) = -10.64501703\n",
      "Iteration  200: Average log likelihood (of data points in batch [20000:20100]) = -4.03616276\n",
      "Iteration  300: Average log likelihood (of data points in batch [30000:30100]) = -14.76470374\n",
      "Iteration  400: Average log likelihood (of data points in batch [40000:40100]) = -4.56769720\n",
      "Iteration  500: Average log likelihood (of data points in batch [02300:02400]) = -1.71720381\n",
      "Iteration  600: Average log likelihood (of data points in batch [12300:12400]) = -9.76736256\n",
      "Iteration  700: Average log likelihood (of data points in batch [22300:22400]) = -9.84104047\n",
      "Iteration  800: Average log likelihood (of data points in batch [32300:32400]) = -5.02920178\n",
      "Iteration  900: Average log likelihood (of data points in batch [42300:42400]) = -6.31326243\n",
      "Iteration 1000: Average log likelihood (of data points in batch [04600:04700]) = -3.69306604\n",
      "Iteration 2000: Average log likelihood (of data points in batch [09200:09300]) = -3.98702992\n",
      "Iteration 3000: Average log likelihood (of data points in batch [13800:13900]) = -2.02646335\n",
      "Iteration 4000: Average log likelihood (of data points in batch [18400:18500]) = -31.03123575\n",
      "Iteration 4769: Average log likelihood (of data points in batch [47600:47700]) = -26.90314015\n"
     ]
    }
   ],
   "source": [
    "batch_size = 100\n",
    "num_passes = 10\n",
    "num_iterations = num_passes * int(len(feature_matrix_train)/batch_size)\n",
    "\n",
    "coefficients_sgd = {}\n",
    "log_likelihood_sgd = {}\n",
    "for step_size in np.logspace(-4, 2, num=7):\n",
    "    coefficients_sgd[step_size], log_likelihood_sgd[step_size] = logistic_regression_SG(feature_matrix_train, sentiment_train,\n",
    "                                       initial_coefficients=np.zeros(194),\n",
    "                                       step_size=step_size, batch_size=100, max_iter=num_iterations)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plotting the log likelihood as a function of passes for each step size\n",
    "\n",
    "Now, we will plot the change in log likelihood using the `make_plot` for each of the following values of `step_size`:\n",
    "\n",
    "* `step_size = 1e-4`\n",
    "* `step_size = 1e-3`\n",
    "* `step_size = 1e-2`\n",
    "* `step_size = 1e-1`\n",
    "* `step_size = 1e0`\n",
    "* `step_size = 1e1`\n",
    "* `step_size = 1e2`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For consistency, we again apply `smoothing_window=30`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2cAAAHVCAYAAAB11oMyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x122e98810>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for step_size in np.logspace(-4, 2, num=7):\n",
    "    make_plot(log_likelihood_sgd[step_size], len_data=len(train_data), batch_size=100,\n",
    "              smoothing_window=30, label='step_size=%.1e'%step_size)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let us remove the step size `step_size = 1e2` and plot the rest of the curves."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2cAAAHVCAYAAAB11oMyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMcYYYwqA/w/8R8ypmN8nTAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x12677b490>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for step_size in np.logspace(-4, 2, num=7)[0:6]:\n",
    "    make_plot(log_likelihood_sgd[step_size], len_data=len(train_data), batch_size=100,\n",
    "              smoothing_window=30, label='step_size=%.1e'%step_size)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [default]",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}