Machine_learning_Algos.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "# Mechanics\n",
    "\n",
    "In this project, we'll learn about the machine learning classifiers.\n",
    "\n",
    "**What is a Classifier?**\n",
    "\n",
    "\n",
    "    \"An algorithm that implements classification, especially in a concrete implementation, is known as a classifier. \n",
    "    The term \"classifier\" sometimes also refers to the mathematical function, implemented by a classification algorithm,that maps input data to a category.\"\n",
    "\n",
    "### Advantages of some particular algorithms\n",
    "\n",
    "Advantages of Naive Bayes: Super simple, you’re just doing a bunch of counts. If the NB conditional independence assumption actually holds, a Naive Bayes classifier will converge quicker than discriminative models like logistic regression, so you need less training data. And even if the NB assumption doesn’t hold, a NB classifier still often does a great job in practice. A good bet if want something fast and easy that performs pretty well. Its main disadvantage is that it can’t learn interactions between features (e.g., it can’t learn that although you love movies with Brad Pitt and Tom Cruise, you hate movies where they’re together).\n",
    "\n",
    "**Advantages of Logistic Regression:** Lots of ways to regularize your model, and you don’t have to worry as much about your features being correlated, like you do in Naive Bayes. You also have a nice probabilistic interpretation, unlike decision trees or SVMs, and you can easily update your model to take in new data (using an online gradient descent method), again unlike decision trees or SVMs. Use it if you want a probabilistic framework (e.g., to easily adjust classification thresholds, to say when you’re unsure, or to get confidence intervals) or if you expect to receive more training data in the future that you want to be able to quickly incorporate into your model.\n",
    "\n",
    "**Advantages of Decision Trees:** Easy to interpret and explain (for some people – I’m not sure I fall into this camp). They easily handle feature interactions and they’re non-parametric, so you don’t have to worry about outliers or whether the data is linearly separable (e.g., decision trees easily take care of cases where you have class A at the low end of some feature x, class B in the mid-range of feature x, and A again at the high end). One disadvantage is that they don’t support online learning, so you have to rebuild your tree when new examples come on. Another disadvantage is that they easily overfit, but that’s where ensemble methods like random forests (or boosted trees) come in. Plus, random forests are often the winner for lots of problems in classification (usually slightly ahead of SVMs, I believe), they’re fast and scalable, and you don’t have to worry about tuning a bunch of parameters like you do with SVMs, so they seem to be quite popular these days.\n",
    "\n",
    "**Advantages of SVMs:** High accuracy, nice theoretical guarantees regarding overfitting, and with an appropriate kernel they can work well even if you’re data isn’t linearly separable in the base feature space. Especially popular in text classification problems where very high-dimensional spaces are the norm. Memory-intensive, hard to interpret, and kind of annoying to run and tune, though, so I think random forests are starting to steal the crown.\n",
    "\n",
    "**But…**\n",
    "\n",
    "Recall, though, that better data often beats better algorithms, and designing good features goes a long way. And if you have a huge dataset, then whichever classification algorithm you use might not matter so much in terms of classification performance (so choose your algorithm based on speed or ease of use instead).\n",
    "\n",
    "And to reiterate what I said above, if you really care about accuracy, you should definitely try a bunch of different classifiers and select the best one by cross-validation. Or, to take a lesson from the Netflix Prize (and Middle Earth), just use an ensemble method to choose them all.\n",
    "\n",
    "#### Dead Kernal\n",
    "In case kernel does not connect, and shows a dead kernal meggese in read : \n",
    "- Go to Kernal -> Change kernal, and choose any of the available kernels.\n",
    "- Press ctrl + s and close the tab.\n",
    "- Open the notebook again and the kernel will connect. If it still does not connect then go to Kernel -> restart.\n",
    "\n",
    "**Note : ** Accuracies may slightly vary due to randomness in train-text split. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Supervised Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "In this project, we will apply all the machine learning classifiers on the given dataset. We will perform exploratory data analysis, visualisation and, finally, we learn how to perform classification tasks using Naive Bayes,Decision Trees, K-Nearest Neighbors, Linear SVC , Support Vector Classifier, Random Forest, Gradeint Decent, and Ada Boost Classifier,etc. For this, we will use some python packages and scikit-learn package (sklearn): a machine learning library for Python which works with numpy array, and pandas DataFrame objects.\n",
    "\n",
    "Please Note: If you get confused, refer to the documentation and just ctrl+f for the object concerned:\n",
    "\n",
    "- Scikit-learn API documentation\n",
    "- Seaborn API documentation\n",
    "- Matplotlib Pyplot documentation\n",
    "- Pandas API documentation\n",
    "- Numpy documentation\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Imports\n",
    "In this project, you are asked to import all the packages and modules you will need. Include all required imports and execute the cell below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/shan/anaconda2/lib/python2.7/site-packages/sklearn/grid_search.py:43: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. This module will be removed in 0.20.\n",
      "  DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.naive_bayes import GaussianNB, MultinomialNB\n",
    "import pandas as pd\n",
    "from collections import defaultdict\n",
    "import numpy as np\n",
    "from sklearn import tree\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy.stats\n",
    "import seaborn as sns\n",
    "import matplotlib\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn import svm\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import confusion_matrix, log_loss\n",
    "\n",
    "from sklearn.utils import resample\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "from sklearn.cluster import MiniBatchKMeans\n",
    "from sklearn import grid_search\n",
    "from sklearn.pipeline import Pipeline\n",
    "\n",
    "from sklearn.kernel_approximation import RBFSampler\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "#### Clarification:\n",
    "\n",
    "    The %matplotlib inline command is a special ipython built in magic command which forces the matplotlib plots to be rendered within the notebook.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Description of the datasets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "The Dataset is based on 4 small preprocessed datasets:\n",
    "It contains 2225 documents\n",
    "\n",
    "**classes.txt:** document id and the class id\n",
    "\n",
    "**docs.txt: ** Description of the classes\n",
    "\n",
    "**mtx.mtx:** \n",
    "It has three columns:\n",
    "\n",
    "    column 0: the id of the term\n",
    "\n",
    "    column 1: the id of the document\n",
    "\n",
    "    column 2: counts/frequency\n",
    "\n",
    "**terms.txt:** each row corresponds to a word/term (id is obviously the row number)\n",
    "\n",
    "**Note: **Execute the cell below to load the csv into in a pandas DataFrame object.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### Load Dataset in Pandas DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Loading dataset\n",
    "\n",
    "#Column names for our dataset \n",
    "column_names = ['term_id','doc_id', 'counts']\n",
    "\n",
    "#Reading Dataset into Pandas Dataframe\n",
    "mtx_data = pd.read_csv( \"./dataset/mtx.mtx\", sep=\"\\s+\", names=column_names)\n",
    "terms_data    = pd.read_csv(\"./dataset/terms.txt\", sep=\"\\s+\", names = ['words'], encoding='utf-8')\n",
    "classes_data  = pd.read_csv( \"./dataset/classes.txt\", sep=\"\\s+\", names=list(['doc_id', 'class_id']))\n",
    "Columns = np.array(list(terms_data.words))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get Your Data in numpy arrays\n",
    "\n",
    "**Note: ** This cell takes time so run the cell below if you have already saved data into ** Data_set.csv**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "\"\"\" Do NOT Run this Cell if data is already loaded in 'Data_set.csv' \"\"\"\n",
    "\n",
    "data_array = np.asarray(mtx_data[mtx_data.columns])\n",
    "classes_array = np.asarray(classes_data[classes_data.columns])\n",
    "\n",
    "# array of zeros\n",
    "zero_arr = np.zeros((2225,1), dtype=int)\n",
    "i=0\n",
    "for col in Columns:\n",
    "    classes_data[col] = zero_arr\n",
    "    if i==2225:\n",
    "        break\n",
    "    i += 1\n",
    "\n",
    "for i in range(classes_array.shape[0]):\n",
    "    \n",
    "    #if (i+1)%10 == 0:\n",
    "    print \"i:\",i,\n",
    "        #break\n",
    "    \n",
    "    doc_id = classes_array[i, 0]        # Getting Document id\n",
    "    mask = data_array[:,1]==doc_id      # Getting indices where condition holds \n",
    "                                        # mask is boolean array\n",
    "    if sum(mask) > 0:                   # Not empty list returned\n",
    "        \n",
    "        terms_ids = data_array[mask, 0]\n",
    "        arr = np.array(terms_ids, dtype=int)\n",
    "        words = list(Columns[list(arr-1)])\n",
    "        #l += len(words)\n",
    "        for word in words:              #Placing 1 on the each occurance of each word in each document\n",
    "            \n",
    "            #classes_data.set_value(i, unicode(word), 1)\n",
    "            \n",
    "            # some characters do not lie in ascii(128) so need to cast them in unicode\n",
    "            classes_data.ix[i][unicode(word)]= 1 \n",
    "\n",
    "print \"No.of NaN Entries:\",classes_data.isnull().sum().sum()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading data from .csv file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "classes_data  = pd.read_csv( \"./dataset/Data_set.csv\", sep=\"\\t\", names=Columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>doc_id</th>\n",
       "      <th>class_id</th>\n",
       "      <th>ad</th>\n",
       "      <th>sale</th>\n",
       "      <th>boost</th>\n",
       "      <th>time</th>\n",
       "      <th>warner</th>\n",
       "      <th>profit</th>\n",
       "      <th>quarterli</th>\n",
       "      <th>media</th>\n",
       "      <th>...</th>\n",
       "      <th>wholli</th>\n",
       "      <th>exploit</th>\n",
       "      <th>commerci</th>\n",
       "      <th>synergi</th>\n",
       "      <th>outright</th>\n",
       "      <th>evid</th>\n",
       "      <th>silvio</th>\n",
       "      <th>berlusconi</th>\n",
       "      <th>sputter</th>\n",
       "      <th>rome</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 2228 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   doc_id  class_id  ad  sale  boost  time  warner  profit  quarterli  media  \\\n",
       "0       0         0   1     0      0     0       0       0          0      0   \n",
       "1       1         0   1     1      1     1       1       1          1      1   \n",
       "2       2         0   0     0      1     1       0       0          0      0   \n",
       "3       3         0   0     1      0     0       0       0          0      0   \n",
       "4       4         0   0     1      0     0       0       1          1      0   \n",
       "5       5         0   0     0      0     1       0       0          0      0   \n",
       "6       6         0   0     0      0     0       0       0          0      0   \n",
       "7       7         0   1     0      1     0       0       0          0      0   \n",
       "8       8         0   0     0      0     0       0       0          0      0   \n",
       "9       9         0   0     0      0     0       0       0          0      0   \n",
       "\n",
       "   ...   wholli  exploit  commerci  synergi  outright  evid  silvio  \\\n",
       "0  ...        0        0         0        0         0     0       0   \n",
       "1  ...        0        0         0        0         0     0       0   \n",
       "2  ...        0        0         0        0         0     0       0   \n",
       "3  ...        0        0         0        0         0     0       0   \n",
       "4  ...        0        0         0        0         0     0       0   \n",
       "5  ...        0        0         0        0         0     0       0   \n",
       "6  ...        0        0         0        0         0     0       0   \n",
       "7  ...        0        0         0        0         0     0       0   \n",
       "8  ...        0        0         0        0         0     0       0   \n",
       "9  ...        0        0         0        0         0     0       0   \n",
       "\n",
       "   berlusconi  sputter  rome  \n",
       "0           0        0     0  \n",
       "1           0        0     0  \n",
       "2           0        0     0  \n",
       "3           0        0     0  \n",
       "4           0        0     0  \n",
       "5           0        0     0  \n",
       "6           0        0     0  \n",
       "7           0        0     0  \n",
       "8           0        0     0  \n",
       "9           0        0     0  \n",
       "\n",
       "[10 rows x 2228 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes_data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Loading Pandas DataFrames into numpy arrays"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Saving DataFrame to .CSV File"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "classes_data.to_csv(\"./dataset/Data_set.csv\", sep=\"\\t\", encoding='utf-8')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have now loaded the data. Let's get a feeling of what the data looks like by using the head() method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>doc_id</th>\n",
       "      <th>class_id</th>\n",
       "      <th>ad</th>\n",
       "      <th>sale</th>\n",
       "      <th>boost</th>\n",
       "      <th>time</th>\n",
       "      <th>warner</th>\n",
       "      <th>profit</th>\n",
       "      <th>quarterli</th>\n",
       "      <th>media</th>\n",
       "      <th>...</th>\n",
       "      <th>wholli</th>\n",
       "      <th>exploit</th>\n",
       "      <th>commerci</th>\n",
       "      <th>synergi</th>\n",
       "      <th>outright</th>\n",
       "      <th>evid</th>\n",
       "      <th>silvio</th>\n",
       "      <th>berlusconi</th>\n",
       "      <th>sputter</th>\n",
       "      <th>rome</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 2228 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   doc_id  class_id  ad  sale  boost  time  warner  profit  quarterli  media  \\\n",
       "0       0         0   1     0      0     0       0       0          0      0   \n",
       "1       1         0   1     1      1     1       1       1          1      1   \n",
       "2       2         0   0     0      1     1       0       0          0      0   \n",
       "3       3         0   0     1      0     0       0       0          0      0   \n",
       "4       4         0   0     1      0     0       0       1          1      0   \n",
       "\n",
       "   ...   wholli  exploit  commerci  synergi  outright  evid  silvio  \\\n",
       "0  ...        0        0         0        0         0     0       0   \n",
       "1  ...        0        0         0        0         0     0       0   \n",
       "2  ...        0        0         0        0         0     0       0   \n",
       "3  ...        0        0         0        0         0     0       0   \n",
       "4  ...        0        0         0        0         0     0       0   \n",
       "\n",
       "   berlusconi  sputter  rome  \n",
       "0           0        0     0  \n",
       "1           0        0     0  \n",
       "2           0        0     0  \n",
       "3           0        0     0  \n",
       "4           0        0     0  \n",
       "\n",
       "[5 rows x 2228 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes_data.head(5) # Display the 5 first rows of the dataframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In Exploratory Data Analysis, we plot graphs and charts to visualize how our data looks like. This is practically very helpful to get understanding about the data, relation among different features, how they affect each other, and which attributes are vital to build a learning model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Histogram"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A histogram is a plot that lets you discover, and show, the underlying frequency distribution (shape) of a set of continuous data. This allows the inspection of the data for its underlying distribution (e.g., normal distribution), outliers, skewness, etc. An example of a histogram, and the raw data it was constructed from, is shown below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhsAAAF+CAYAAAAr9tjJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzt3X+UZGV95/F3TXUPM0TRzkwLwagYcwJGE3VGNKsbY8Bd\nA4eJycb97lnHGJYgMYgwioj4I5BEd1TEmKggAiohMPEbk7MSMGIWkrgcXWSc5UBIJMYEBxChZ2h+\nTBhmupraP24Ndpr+Ud1dT1d11ft1zpyqus+9t771cLn9qef+qFqz2USSJKmUVd0uQJIk9TfDhiRJ\nKsqwIUmSijJsSJKkogwbkiSpKMOGJEkqyrAhSZKKMmxIkqSiDBuSJKkow4YkSSpqqNsFTBUR7wXe\nCjwV+Abw5sz8XkQcA2wFjgJ2Alsz86opy50OnAocBtwKbMnMHctdvyRJerJar/w2SkS8lSowvA74\nAfCBVtOHgO8ApwHbgJ8HrgZ+PjN3RMQm4PPAa4HbgDOALcDzMnPvcn4GSZL0ZL00svEO4B2Z+c+t\n11sAIuJM4I7MvLw1/fqIuBo4mSqcnAJ8LjO3t+Y/nypwbAJyGeuXJEkz6ImwERGHA88F1kXE7cCh\nwA1UYWIjMP2QyA4gWs83Uo14AJCZzYi4BTgaw4YkSV3XE2ED+PHW4+uBY4A68OfAJcDBwF3T5n8A\nWN96vg4Yn6N9PuuoDsHcCTy2kKIlSRpwa4AjgOuA3bPN1Ctho9Z6/HBm3gcQEecCfwX89ZT2+ZZf\njNcCVy5heUmSBt1m4KrZGnslbPyg9fjQlGl3UoWIYarRh6nWAfe3no/N0n5bm+99J8DDDz/M5ORk\nm4tIkqR6vc4hhxwCrb+ls+mVsHE38DDwYuCW1rTnAvuBLwNvmjb/0cBNrefbqc7buAIgIlYBG4BL\n23zvxwD27dvHxMTEIsuXJGnwDA8PH3g652kIPRE2MnMyIi4D3hsR/wd4BHg/VYD4Y+D9EXES1eGO\nY4HjgJe3Fr8I2BYR26jusXEW1Ye+dnk/hSRJmkkv3UH0HOArwDep7qtxB3BGZo4BJwBvAx4ELgA2\nZ+btAJl5XWvZpDo55Vjg+Mzct+yfQJIkPUnP3NSrizYA3xobG/MwiiRJCzA8PMzo6CjMfJuKJ/TS\nyIYkSepDhg1JklSUYUOSJBVl2JAkSUUZNiRJUlGGDUmSVJRhQ5IkFWXYkCRJRRk2JElSUYYNSZJU\nlGFDkiQVZdiQJElFGTYkSVJRhg1JklSUYUOSJBVl2JAkSUUZNiRJUlGGDUmSVJRhQ5IkFWXYkCRJ\nRRk2JElSUYYNSZJUlGFDkiQVZdiQJElFGTYkSVJRhg1JklSUYUOSJBVl2JAkSUUZNiRJUlGGDUmS\nVJRhQ5IkFWXYkCRJRRk2JElSUYYNSZJUlGFDkiQVZdiQJElFGTYkSVJRhg1JklSUYUOSJBVl2JAk\nSUUNdbsASRpkExMT7N+/v9tlrAirV69meHi422VoEXombETE48A+oAnUWo+XZOYZEXEMsBU4CtgJ\nbM3Mq6YsezpwKnAYcCuwJTN3dLK++r69NPfu7eQq+1Zt7VomD1rb7TKkFWH//v1cdtll3S5jRfjN\n3/xNw8YK1TNhgypc/FRm3jV1YkQcBnwJOA3YBvw8cHVEfDszd0TEJuBc4LXAbcAZwDUR8bzM7Fg6\naO7dyz0ffFenVtfXnvnej4BhQ5LU0kvnbNRa/6bbDNyRmZdn5v7MvB64Gji51X4K8LnM3J6Z+4Dz\nqYLLpuUoWpIkza2XRjYAPhwRrwAOAb4AnAlsBKYfEtkBROv5RqoRDwAysxkRtwBHA1m8YkmSNKde\nChvfAL4KvAn4CaqwcSGwDrhr2rwPAOtbz9cB43O0t2VoaO6umKzNNOiiGdVqHleV2lRz39K2mvuW\nnjPf384n5itcR9sy85VTXt4REe8G/hL4GjMfXplqyf+3joyMzNm+a3z3Ut9iYAzV66wfHe12GdKK\nsGfPnm6XsGLU63VG3besSD0TNmZwJ1AHHqcavZhqHXB/6/nYLO23LeTNxsfHaTQas7ZPTk4uZHUD\nrTE5ydjYWLfLkFYE9y3tm3Tf0nOGhobm/bIOPRI2IuLFwBsz851TJv808BjwZeDEaYscDdzUer6d\n6ryNK1rrWgVsAC5dSA2NRoOJiYlZ21c1mwtZ3WBrNufsS0k/1HTf0ram+5YVqyfCBtUoxSkRcT/w\nceAI4PeAi4E/Ac6NiJOAK4FjgeOAl7eWvQjYFhHbqO6xcRZVSLl2OT+AJEmaWU9c+pqZ3weOB14H\n7AJupBrRODszx4ATgLcBDwIXAJsz8/bWstcB51BdebKbKowc37oMVpIkdVmvjGyQmTcCr5yj7SVz\nLHsx1SiIJEnqMT0xsiFJkvqXYUOSJBVl2JAkSUUZNiRJUlGGDUmSVJRhQ5IkFWXYkCRJRRk2JElS\nUYYNSZJUlGFDkiQVZdiQJElFGTYkSVJRhg1JklSUYUOSJBVl2JAkSUUZNiRJUlGGDUmSVJRhQ5Ik\nFWXYkCRJRRk2JElSUYYNSZJUlGFDkiQVZdiQJElFGTYkSVJRhg1JklSUYUOSJBVl2JAkSUUZNiRJ\nUlGGDUmSVJRhQ5IkFWXYkCRJRRk2JElSUYYNSZJUlGFDkiQVZdiQJElFGTYkSVJRhg1JklTUULcL\nkNQ7mo/XaTSa3S5jRRgaqlFbNdntMqQVwbAh6QmNRpMbvnJPt8tYEY75pWcyvLrbVUgrg4dRJElS\nUT03shERfwCckZmrWq+PAbYCRwE7ga2ZedWU+U8HTgUOA24FtmTmjmUvXJIkzainRjYi4sXArwPN\n1usfA74EXAiMAluASyJiQ6t9E3Au8EbgUOAa4JqIWLv81UuSpJn0TNiIiBpwEXDBlMmbgTsy8/LM\n3J+Z1wNXAye32k8BPpeZ2zNzH3A+VVDZtIylS5KkOfRM2ADeAuwFrpoybQMw/ZDIDuDo1vONU9sz\nswncMqVdkiR1WU+csxERhwLnAa+a1rQOuGvatAeA9VPax+dolyRJXdYTYYPq0MllmXlHRDxnWltt\nnmXna2/L0NDcXTFZ68jbDIZajeHh4W5XoUWYmPC+EW2r0ZHtvOa+pW21Du1bVj2+Fyb3daCiAVA/\niMdXzX4a5Hx/O5+Yr1P1LFZEHAu8Anhza9LU//PGqEYvploH3D9P+20LrWNkZGTO9l3juxe6yoE1\nVK+zfnS022VoEe65e1e3S1gxhupDjI4ufRB1z549HahmMNTrdUY7sG958L5/4bF//HQHKup/a174\nVtZ1oM+7HjaoTgJ9BrAzIqA6j6QWEfdTjXi8Ydr8RwM3tZ5vpzpv4wqAiFhFdZ7HpQstYnx8nEaj\nMWv75KTf+NrVmJxkbGys22VoERpu521rTDY6sp27b2nfZIf2Lavs87bN1+dDQ0PzflmH3ggbbwfe\nN+X1s4BvAC+iqu+ciDgJuBI4FjgOeHlr3ouAbRGxjeoeG2cBjwHXLrSIRqPBxMTErO2rmt7CuW3N\n5px9qR7W7KVzxntck45s5033LW1rdmjfsto+b1un+rzrYSMzHwIeOvA6IoaBZmbe23p9AvAJ4FPA\nncDmzLy9tex1EXEOkFT34bgZOL51GawkSeoBXQ8b02Xm94D6lNc3Ai+ZY/6LgYuXoTRJkrQIjplK\nkqSiem5kQzrg0UnYu98TudqxdnWdg+vzzydJ3WDYUM/au3+SM/Ob3S5jRbggXsbBa00bknqTh1Ek\nSVJRhg1JklSUYUOSJBVl2JAkSUUZNiRJUlGGDUmSVJRhQ5IkFWXYkCRJRRk2JElSUYYNSZJUlGFD\nkiQVZdiQJElFGTYkSVJRhg1JklSUYUOSJBVl2JAkSUUZNiRJUlGGDUmSVJRhQ5IkFWXYkCRJRRk2\nJElSUYYNSZJUlGFDkiQVZdiQJElFGTYkSVJRiwobEfHZiHjqLG0/FRFfXFpZkiSpXyx2ZOM3gDWz\ntP00sGmR65UkSX1maCEzR8TjQLP18gcRMdusty2lKEmS1D8WFDaAnwBeCVwBfBL4txnmGQeuXGJd\nkiSpTywobGTmncCdEfGTwEczc6awIUmS9ISFjmwAkJm/CxARTwFGgNoM8+xcWmmSJKkfLCpsRMRL\ngT8GjpxjtvqiKpIkSX1lUWED+BQwDJwHjPHDk0YlSZL+ncWGjRcAr87M7Z0sRpIk9Z/F3mfjPma+\nEkWSJOnfWWzY+ChwdkQsdmREkiQNiMWGhZ8CXkZ1GewOYM+09mZmbl5SZZIkqS8sNmz8SutxAviZ\nGdo9YVSSJAGLv8/GcztdSES8CLgAeCmwF/g74PTMvD8ijgG2AkcBO4GtmXnVlGVPB04FDgNuBbZk\n5o5O1yhJkhauJ35iPiJWA9cBNwCjwAuBQ4GLIuIw4EvAha22LcAlEbGhtewm4Fzgja1lrgGuiYi1\ny/05JEnSky32pl73Ms+hksw8fAGrPBh4D/D5zHwc2B0RfwGcBmwG7sjMy1vzXh8RVwMnU41mnAJ8\n7sBluBFxPnAG1S/P5gJqkCRJBSz2nI3reHLYeArVIZD9wJ8vZGWZ+SDw2QOvI+JI4ETgC8BGYPoh\nkR3AgZ+c3Qhsm7KuZkTcAhyNYUOSpK5b7DkbJ840PSLqwCXA/YtZb0Q8G/gO1a3OP0N1h9K/Au6a\nNusDwPrW83VUvzQ7W3tbhobm7orJ2pN+/kWzqdUYHh5e+moee7wDxQyGWof6fGJisgPVDIgandnO\n3be0rVPbeW3CPm/XfH0+39/OJ+brVEEAmTkZER+lCggfX8TyO4GDIuJ5VGHjilbTfFvGkreckZGR\nOdt3je9e6lsMjKF6nfWjo0tez65H7+1ANYOhXq8z2oE+v+fuXR2oZjAM1YcYHV3Qd5oZ7dkz/c4B\nmk2ntvMH73uEiQ7UMwjq9TrrOtDnJW7KdTALHFWYLjO/GxHvBb4OXEs1ejHVOn44ejI2S/ttC3nP\n8fFxGo3GrO2Tk37ja1djcpKxsbElr8c+b99kh/q8YZ+3rTHZcDtfZp3azlfZ522br8+Hhobm/bIO\niz9B9JQZJteorhZ5E/CPC1zfLwIXZeZRUyY3W/++Cbx+2iJHAze1nm+nOm/jita6VgEbgEsXUkOj\n0WBiYvasu6rprUPa1mzO2Zftr8Y+b1ezQ31OsycuUFsZmridL7NObeer7fO2darPFzuy8ek52v4J\n+O0Fru9bwCER8WGq8zSeQnU569eAi4AzI+Ik4ErgWOA44OWtZS8CtkXENqp7bJwFPEY1IiJJkrps\nsV9jnjvDvyOAp2fmUZl580JWlpkPA/+J6hboY1SHQB4E3pCZu4ATgLe1pl0AbM7M21vLXgecQ3Xl\nyW6qMHJ8Zu5b5GeTJEkdtNirUb7X6UJa4eEXZ2m7EXjJHMteDFzc6ZokSdLSLfoE0dadO99GdX7E\nU6lGHW4GPpKZX+tMeZIkaaVb1GGUiPhvVLcQPwz4IvCHwP+iOpRyfUT8504VKEmSVrbFjmy8C/ij\nzNwyvSEiLqY6ufOrSylMkiT1h8WGjecDb5il7Q+pDqdIkiQt+mqUBrBmjnV6n2lJkgQsPmx8Ezg3\nIv7dDdNbPxV/XqtdkiRp0YdR3g9cD9wbETcDDwFPo7qz5xqqe11IkiQtbmQjM79BdYvwLwLPaD1f\nT/VT7xsy86Y5FpckSQNkKT/E9gvAv2bmWw5MiIjLgNdQ3bJckiRp0ffZeBvwCZ4cVu4BPhYRb11q\nYZIkqT8s9gTRU4FTMvODUydm5u8ApwCnLbUwSZLUHxYbNp4D/O0sbX/bapckSVp02LgbeMUsba8B\nvr/I9UqSpD6z2BNELwY+HREbqO6p8QgwQnXS6G9QXRorSZK06LDxMeAg4B3A26dMfwj4QGZ+ZKmF\nSZKk/rCosJGZTeB/RsT5wPOApwP3Azszs9HB+iRJ0gq3lPtskJkTwLc7VIskSepDiz1BVJIkqS2G\nDUmSVJRhQ5IkFWXYkCRJRRk2JElSUYYNSZJUlGFDkiQVZdiQJElFGTYkSVJRhg1JklSUYUOSJBVl\n2JAkSUUZNiRJUlGGDUmSVJRhQ5IkFWXYkCRJRRk2JElSUYYNSZJUlGFDkiQVZdiQJElFGTYkSVJR\nhg1JklSUYUOSJBVl2JAkSUUNdbuAAyLi2cDHgVcBE8BXgDMy8+GIOAbYChwF7AS2ZuZVU5Y9HTgV\nOAy4FdiSmTuW+SNIkqQZ9NLIxl8CDwDPAjYCLwA+GhGHAV8CLgRGgS3AJRGxASAiNgHnAm8EDgWu\nAa6JiLXL/gkkSdKT9ETYiIinATcD52Tm3sz8PnA51SjHZuCOzLw8M/dn5vXA1cDJrcVPAT6Xmdsz\ncx9wPtAENi37B5EkSU/SE4dRMvMhfhgeDngWcA/VKMf0QyI7gGg93whsm7KuZkTcAhwNZJGCJUlS\n23oibEwXES8FTgN+GTgbuGvaLA8A61vP1wHjc7S3ZWho7q6YrNUWsrrBVqsxPDy89NU89ngHihkM\ntQ71+cTEZAeqGRA1OrOdu29pW6e289qEfd6u+fp8vr+dT8zXqYI6JSJeSXWY5OzMvCEizgbm2zKW\nvOWMjIzM2b5rfPdS32JgDNXrrB8dXfJ6dj16bweqGQz1ep3RDvT5PXfv6kA1g2GoPsTo6IK+08xo\nz549HahmMHRqO3/wvkeY6EA9g6Ber7OuA33eU2GjdbLnFcBbM/PK1uQxqtGLqdYB98/TfttC3nt8\nfJxGozFr++Sk3/ja1ZicZGxsbMnrsc/bN9mhPm/Y521rTDbczpdZp7bzVfZ52+br86GhoXm/rEMP\nhY2IeAXweeDXWieBHrAdOHHa7EcDN01p30gVUoiIVcAG4NKFvH+j0WBiYvasu6rZXMjqBluzOWdf\ntr8a+7xdzQ71Oc2eOGd8ZWjidr7MOrWdr7bP29apPu+JsBERdeASqkMn109rvhI4LyJOaj0/FjgO\neHmr/SJgW0Rso7rHxlnAY8C1y1G7JEmaW0+EDeA/UN2w648i4hNUl67WWo9HAicAnwA+BdwJbM7M\n2wEy87qIOIfqypNRqktoj29dBitJkrqsJ8JGZt4I1OeY5S7gJXMsfzFwcafrkiRJS+cBWkmSVJRh\nQ5IkFWXYkCRJRRk2JElSUYYNSZJUlGFDkiQVZdiQJElFGTYkSVJRhg1JklSUYUOSJBVl2JAkSUUZ\nNiRJUlGGDUmSVJRhQ5IkFWXYkCRJRRk2JElSUYYNSZJUlGFDkiQVZdiQJElFGTYkSVJRhg1JklSU\nYUOSJBVl2JAkSUUZNiRJUlGGDUmSVJRhQ5IkFWXYkCRJRRk2JElSUYYNSZJUlGFDkiQVZdiQJElF\nGTYkSVJRhg1JklSUYUOSJBVl2JAkSUUZNiRJUlGGDUmSVJRhQ5IkFWXYkCRJRRk2JElSUUPdLuCA\niHgtcDlwQ2a+YVrbMcBW4ChgJ7A1M6+a0n46cCpwGHArsCUzdyxX7ZIkaXY9MbIREWcBHwf+aYa2\nw4AvARcCo8AW4JKI2NBq3wScC7wROBS4BrgmItYuT/WSJGkuPRE2gL3Ay4DvztC2GbgjMy/PzP2Z\neT1wNXByq/0U4HOZuT0z9wHnA01g0zLULUmS5tETYSMzP5mZj8zSvBGYfkhkB3D0TO2Z2QRumdIu\nSZK6qCfCxjzWAePTpj0ArG+zXZIkdVHPnCA6j9oS2+c1NDR3V0zWlvwWg6NWY3h4eOmreezxDhQz\nGGod6vOJickOVDMganRmO3ff0rZObee1Cfu8XfP1+Xx/O5+Yr1MFFTRGNXox1Trg/nnab1vIm4yM\njMzZvmt890JWN9CG6nXWj44ueT27Hr23A9UMhnq9zmgH+vyeu3d1oJrBMFQfYnR06QOoe/bs6UA1\ng6FT2/mD9z3CRAfqGQT1ep11HejzlRA2tgMnTpt2NHDTlPaNwBUAEbEK2ABcupA3GR8fp9FozNo+\nOek3vnY1JicZGxtb8nrs8/ZNdqjPG/Z52xqTDbfzZdap7XyVfd62+fp8aGho3i/rsDLCxpXAeRFx\nUuv5scBxwMtb7RcB2yJiG9U9Ns4CHgOuXcibNBoNJiZmz7qrms2FVz6oms05+7L91djn7Wp2qM9p\nroTTuHpEE7fzZdap7Xy1fd62TvV5T+xZImJvRDxKda+M/zrlNZk5BpwAvA14ELgA2JyZt7farwPO\nARLYTRVGjm9dBitJkrqsJ0Y2MnPOG3Bl5o3AS+Zovxi4uNN1SZKkpeuJkQ1JktS/DBuSJKkow4Yk\nSSrKsCFJkooybEiSpKIMG5IkqSjDhiRJKsqwIUmSijJsSJKkogwbkiSpKMOGJEkqyrAhSZKKMmxI\nkqSiDBuSJKkow4YkSSrKsCFJkooybEiSpKIMG5IkqSjDhiRJKsqwIUmSijJsSJKkogwbkiSpKMOG\nJEkqyrAhSZKKMmxIkqSiDBuSJKkow4YkSSrKsCFJkooybEiSpKIMG5IkqSjDhiRJKsqwIUmSijJs\nSJKkogwbkiSpKMOGJEkqyrAhSZKKMmxIkqSiDBuSJKkow4YkSSrKsCFJkooybEiSpKKGul1AJ0TE\ns4ELgZ8DHgG+kJnv7m5VkiQJ+mdk4y+Au4AjgNcAvxoRW7pakSRJAvogbETES4GfBc7OzD2Z+V3g\nY8Ap3a1MkiRBH4QNYANwZ2Y+PGXaDuDIiPiRLtUkSZJa+uGcjXXA+LRpD7Qe1wP/Ns/yawAOOugg\nhoZm747JfWtZe8TzFlvjQBlau5b62rVLXs/ax/fzk4c+vQMV9b+1B61m7drVS15PrTbJuvVL/283\nCFYfNMSaNfUlr2dycpLDDz+8AxX1vzVr1rC2A/uWWqMBT31mByrqf0Or11Afmr3P6/Un/h9YM+d6\nOlhTN9WWsOwRAIcccsjcc42MsP6Dn1jC22ihRkbgwpMO7XYZA+eXX7++2yUMlJGREU4//fRulzFg\nRmD0Hd0uot8cAXx9tsZ+CBtjVKMbU60Dmq22+VwHbAbuBB7raGWSJPW3NVRB47q5ZuqHsLEdeHZE\n/GhmHjh88jLgHzLz0TaW3w1cVaw6SZL626wjGgfUms3mchRSVER8Hfh74EzgmcC1wPmZ+emuFiZJ\nkvriahSA11OFjB8ANwCfN2hIktQb+mJkQ5Ik9a5+GdmQJEk9yrAhSZKKMmxIkqSiDBuSJKkow4Yk\nSSrKsCFJkorqhzuIDpyIeDZwIfBzwCPAFzLz3d2tqv9FxGuBy4EbMvMN3a6n37W2848DrwImgK8A\nZ0z7hWd1UES8CLgAeCmwF/g7qj6/r6uFDYiI+AOq/u67gYC++0AD4i+Au6juR/8a4FcjYktXK+pz\nEXEW1R++f+p2LQPkL6l+wflZwEbgBcBHu1pRH4uI1VS/b3EDMAq8EDiU6ouNCouIFwO/TvW7Xn3H\nsLHCRMRLgZ8Fzs7MPZn5XeBjwCndrazv7aX6zZ3vdruQQRARTwNuBs7JzL2Z+X2qUaVXdbeyvnYw\n8B7gQ5k5kZm7qb7YvLC7ZfW/iKgBF1GNKvUlD6OsPBuAO6cNJe8AjoyIH8nMf+tSXX0tMz8JEBHd\nLmUgZOZDwMnTJj8buKcL5QyEzHwQ+OyB1xFxJHAi8KfdqmmAvIXqC81VwAe6XEsRho2VZx0wPm3a\ngV+7XQ8YNtR3WiN6pwEndLuWftc6V+Y7QB34DHBeVwvqcxFxKFUf9/WonYdRVqZatwuQlktEvJLq\nXIJ3ZebfdLuefpeZOzPzIODI1r8/6XJJ/e4C4LLMvKPbhZRk2Fh5xqhGN6ZaR3VS0djylyOVExGb\ngGuB0zPzU92uZ5C0zgd7L/DfI2L6PkcdEBHHAq8Afr81qW+/SBo2Vp7twLMj4kenTHsZ8A+Z+WiX\napI6LiJeAXwe+LXMvLLL5fS9iPjFiPj2tMnN1r/9XShpEGwGngHsjIgx4FtALSLujz47QcxzNlaY\nzLwlIm4GPhQRZwLPBN4OnN/dyqTOiYg6cAnVVVfXd7ueAfEt4JCI+DDVOQRPAc4FvpaZj3SzsD72\nduB9U14/C/gG8CKefG7eilZrNvvykt6+FhGHU+2IXw08BFyUmb8/50JakojYS/UNb7g1qQE0M/Pg\n7lXVvyLiP1LdUGof1dByc8rjkZl5VxfL61sR8QLgk8DRwB7geuCdmXlvVwsbEBHxHOBfMrPe7Vo6\nzbAhSZKK8pwNSZJUlGFDkiQVZdiQJElFGTYkSVJRhg1JklSUYUOSJBVl2JAkSUUZNiRJUlGGDUmS\nVJS/jSKpKyLiPOB3gDWZ6Q99SX3MkQ1J3XLgF0Ul9TnDhiRJKsrDKJKKiYga8A7gt4AfB74HXJiZ\nn5hh3jrVT5pvpvqp7d3AjVS/Ovq91jyrga3AfwEOAx4ErgPekZkPtOY5BTgN+AlgP3Az8O7M/H/l\nPqmkuTiyIamkc4DzgN8FXgB8BPhYRLxlhnnfA7wLeCfwXOAE4DnAF6fM834ggBOBnwReD7wEuAIg\nIo4FLgTOB54PvAp4APhqRKzp6CeT1DZHNiQVERHDVKMan8nMK1uT/zUiDgeeNsMinwL+NDO/03p9\nT0RcBlwYEesyczewAbg1M/9uyjzHASOt1xuAPcCVmfk4cFdEnAS8EJjs9GeU1B7DhqRSngf8KPDN\nqRMz84MAEXHutPn3AW+KiNcBzwRW88N91DqqwypfAi6KiD8D/gz4m8z8PvD91nx/TXWFy/9tBZUb\nWuHl5g5/NkkL4GEUSaU8vfX4SJvzXwWcCXwG+AXgRVTB4QmZ+RlgE7AW+Cxwb0R8NSKe32q/Bfg5\n4HaqQzd3RMTfR8SmJX4WSUvgyIakUu5vPY7MORcQEU+lOkdja2Z+csr0J+2jMvPLwJdbh2leA3wI\nuJbqhFAy83bgf7SW3wi8G/jziHh+Zn53SZ9I0qIYNiSVshN4iGqU4sA5G0TE71EdJtk5Zd5hoAbs\nmjLfKqorUwBqrStbfgW4OTPvzswJ4K8i4iCqMPF04KeBRmZ+EyAzvxURbwZ+DfgZwLAhdUGt2fSe\nOpLKiIj3A+8Dfhv438CrgYuprjhZT3WYZG1m7o+IO4C9wBupvgh9gCqQ/BbwFmAb1TkZjwNnA3cC\nzwA+CPzgvRdmAAAAuklEQVRYZr44IrZSXalyKvAtYA3w5tb7H5mZ95T+zJKezHM2JBWTmb9Pdbnq\ne4BvU10K+87M/FRrlqnfdt5AdV+Mm4CkuuT1dODrwB9RXeb6OqrRiQT+meqE0V3AL7fW8T7gUqpL\nX79NdZ+OFwG/ZNCQuseRDUmSVJQjG5IkqSjDhiRJKsqwIUmSijJsSJKkogwbkiSpKMOGJEkqyrAh\nSZKKMmxIkqSiDBuSJKkow4YkSSrKsCFJkor6/wD1TqLE/ykjAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcab31608d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.countplot(x='class_id', data=classes_data)\n",
    "ax.set_xticklabels(['0', '1','2','3','4'])\n",
    "ax.set_xlabel('class')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualisation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we want to get a feeling for how the presence or absence of some specific terms could affect the outcome. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Lets Visualize the data\n",
    "\n",
    "**Note: ** Here we are using seaborn pairplot function to see each across all other features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x7febaf815990>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDoAAAPeCAYAAADtX3qjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nERERERERkZqhRIeIiIiIiIiI1Iz/D43/vrJS21YnAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7febaf7c4b90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(classification_data, hue=\"class\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using seaborn, produce one figure with three countplots, one for each of the frequency variables for the words make."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAF+CAYAAABQ0hoFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJBVlGJEkSUUZRiRJUlGGEUmSVNT/B2lojtVIiJOCAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaaf094d50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAF+CAYAAABQ0hoFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkiSpKMOIJEkqyjAiSZKK+n9BjmZoM5SAhQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaaeef4610>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAF+CAYAAABQ0hoFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nlGFEkiQV9f8BXy93bD9ZyIcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaaecbf510>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.countplot(x='boost', data=classes_data)\n",
    "plt.show()\n",
    "\n",
    "ax = sns.countplot(x='time', data=classes_data)\n",
    "plt.show()\n",
    "\n",
    "ax = sns.countplot(x='warner', data=classes_data)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Repeat the above but split the bars showing the proportion of terms that belong to class 0/1/2/3/4. Hint: This only requires you to use the hue input argument to use different colours for the 'class_id' variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhsAAAF+CAYAAAAr9tjJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUqUYUOSJCXKsCFJkhL1/wO+Vp3s778nnQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaaf6aed10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhsAAAF+CAYAAAAr9tjJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhWSXwQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaaf456090>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhsAAAF+CAYAAAAr9tjJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaaf42e7d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Your code goes here\n",
    "ax = sns.countplot(x='ad', hue='class_id', data=classes_data)\n",
    "plt.show()\n",
    "\n",
    "ax = sns.countplot(x='sale', hue='class_id', data=classes_data)\n",
    "plt.show()\n",
    "\n",
    "ax = sns.countplot(x='rome', hue='class_id', data=classes_data)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Applying Classifiers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we are in a position to apply all machine learning classifiers we studied in this course over the cleaned data. As discussed above, we'll prefer to use sklearn library in python which provides us a quality framework to apply classifiers with great efficiency. Lets discuss & apply all classifiers one by one. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Splitting Data into Trainig and Testing parts\n",
    "We are splitting data into testing and training sets.\n",
    "**Training Set:** Used to build the model\n",
    "**Testing Set:**  Used to test the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2225, 2228)\n",
      "(2002, 2227) (2002,) (223, 2227) (223,)\n",
      "False False False False\n"
     ]
    }
   ],
   "source": [
    "arr = np.asarray(classes_data[classes_data.columns])\n",
    "#Check the shape of new array\n",
    "print arr.shape\n",
    "idx = np.arange(1,2228, 1)\n",
    "\n",
    "# I need to replace first index entry '1' with '0' because I need first column not 2nd from array\n",
    "idx[0] = 0\n",
    "\n",
    "# Splitting data into two parts\n",
    "# 90% ---> Training Data\n",
    "# 10% ---> Testing Data\n",
    "Xtrain, Xtest, ytrain, ytest = train_test_split(arr[:, idx],arr[:,1],train_size = 0.9, test_size=0.1, random_state=0)\n",
    "\n",
    "#Print Shape of splitted Data\n",
    "print Xtrain.shape,ytrain.shape,Xtest.shape,ytest.shape                                              \n",
    "\n",
    "# Find if any 'nan' value is present\n",
    "print np.isnan(Xtrain).sum()!=0,np.isnan(Xtest).sum()!=0,np.isnan(ytrain).sum()!=0,np.isnan(ytest).sum()!=0 "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Confusion Matrix Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def print_confusion_matrix(plabels,tlabels):\n",
    "    \"\"\"\n",
    "        functions print the confusion matrix for the different classes\n",
    "        to find the error...\n",
    "        \n",
    "        Input:\n",
    "        -----------\n",
    "        plabels: predicted labels for the classes...\n",
    "        tlabels: true labels for the classes\n",
    "        \n",
    "        code from: http://stackoverflow.com/questions/2148543/how-to-write-a-confusion-matrix-in-python\n",
    "    \"\"\"\n",
    "    plabels = pd.Series(np.squeeze(plabels))\n",
    "    tlabels = pd.Series(np.squeeze(tlabels))\n",
    "    \n",
    "    # draw a cross tabulation...\n",
    "    df_confusion = pd.crosstab(tlabels,plabels, rownames=['Actual'], colnames=['Predicted'], margins=True)\n",
    "    \n",
    "    #print df_confusion\n",
    "    return df_confusion     \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Multinomial Naive Bayes classification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "Given the transformed dataset, we now wish to train a Naïve Bayes classifier to distinguish each term belonging to class (0,1,2,3,4) from other terms in all documents by fitting a distribution of the number of occurrences of each word for all documents. Read about the Naive Bayes classifier and the underlying assumption if you are not already familiar with it. In this project we'll firts focus on the Multinomial Naive Bayes classifier.\n",
    "\n",
    "We will make use of the MultinomialNB class in sklearn. Check out the user guide description and documentation to familiarise yourself with this class.\n",
    "\n",
    "All classifiers in sklearn implement a fit() and predict() method. The first learns the parameters of the model and the latter classifies inputs. For a Naive Bayes classifier, the fit() method takes at least two input arguments X and y, where X are the input features and y are the labels associated with each example in the training dataset (i.e. targets).\n",
    "\n",
    "As a first step we extract the input features and targets from the DataFrame. Now we want to train a Multinomial Naive Bayes classifier. Initialise a MultinomialNB object and fit the classifier using the X and y arrays extracted in the cell above."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MultinomialNB(alpha=1.0, class_prior=None, fit_prior=True)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating Multinomial Naive Bayes Object\n",
    "classifier = MultinomialNB()\n",
    "\n",
    "# Now we are building our model\n",
    "classifier.fit(Xtrain, ytrain)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.955156950673\n",
      "Mean Absolute Error: 0.0627802690583\n"
     ]
    }
   ],
   "source": [
    "# Predicting the Outcome\n",
    "ypredicted = classifier.predict(Xtest)\n",
    "\n",
    "#Calculating Mean Absolute Error which tells us how much our model is misclassifying\n",
    "mae = mean_absolute_error(ytest, ypredicted)\n",
    "\n",
    "print \"Accuracy = \", np.sum(ypredicted==ytest)/float(ytest.shape[0])\n",
    "print \"Mean Absolute Error:\",mae\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>51</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>32</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>47</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>52</td>\n",
       "      <td>34</td>\n",
       "      <td>51</td>\n",
       "      <td>52</td>\n",
       "      <td>34</td>\n",
       "      <td>223</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0   1   2   3   4  All\n",
       "Actual                            \n",
       "0          51   2   3   0   0   56\n",
       "1           1  32   1   0   0   34\n",
       "2           0   0  47   1   1   49\n",
       "3           0   0   0  50   0   50\n",
       "4           0   0   0   1  33   34\n",
       "All        52  34  51  52  34  223"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "print_confusion_matrix(ypredicted.ravel(),ytest.ravel())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Naive Bayes Classifier\n",
    "Naive Bayes, also known as Naive Bayes Classifiers are classifiers with the assumption that features are **statistically independent** of one another. \n",
    "Unlike many other classifiers which assume that, for a given class, there will be some correlation between features, naive Bayes explicitly models the features as conditionally independent given the class. \n",
    "While this may seem an overly simplistic (naive) restriction on the data, in practice naive Bayes is competitive with more sophisticated techniques and enjoys some theoretical support for its efficacy.\n",
    "\n",
    "In spite of the great advances of the Machine Learning in the last years, it has proven to not only be simple but also fast, accurate and reliable. It has been successfully used for many purposes, but it works particularly well with natural language processing (NLP) problems.\n",
    "\n",
    "#### Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GaussianNB(priors=None)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Create a Gaussian Classifier\n",
    "gb = GaussianNB()\n",
    "\n",
    "# Train the model using the training sets \n",
    "gb.fit(Xtrain, ytrain)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.762331838565\n",
      "Mean Absolute Error: 0.515695067265\n"
     ]
    }
   ],
   "source": [
    "\n",
    "#Predict Output \n",
    "ypredicted= gb.predict(Xtest)\n",
    "\n",
    "#Calculating Mean Absolute Error which is how much error is in classification\n",
    "mae = mean_absolute_error(ytest, ypredicted)\n",
    "\n",
    "# Calculating how good is our model is performing on unknown examples\n",
    "print \"Accuracy = \", np.sum(ypredicted==ytest)/float(ytest.shape[0])\n",
    "print \"Mean Absolute Error:\",mae\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>50</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>33</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>42</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>26</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>72</td>\n",
       "      <td>28</td>\n",
       "      <td>39</td>\n",
       "      <td>45</td>\n",
       "      <td>39</td>\n",
       "      <td>223</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0   1   2   3   4  All\n",
       "Actual                            \n",
       "0          50   2   1   0   3   56\n",
       "1           6  19   1   1   7   34\n",
       "2          12   0  33   1   3   49\n",
       "3           1   5   2  42   0   50\n",
       "4           3   2   2   1  26   34\n",
       "All        72  28  39  45  39  223"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print_confusion_matrix(ypredicted.ravel(),ytest.ravel())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Support Vector Classifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**“Support Vector Machine”** (SVM) is a supervised machine learning algorithm which can be used for both classification or regression challenges. \n",
    "However,  it is mostly used in classification problems. In this algorithm, we plot each data item as a point in n-dimensional space (where n is number of features you have) with the value of each feature being the value of a particular coordinate. \n",
    "Then, we perform classification by finding the hyper-plane that differentiate the all classes very well.\n",
    "\n",
    "#### Cross Validation for SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.93573033707865172"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# from sklearn import svm\n",
    "clf = svm.SVC(kernel='linear', C=1).fit(Xtrain, ytrain)\n",
    "# clf.score(Xtest, ytest)  \n",
    "\n",
    "from sklearn.model_selection import cross_val_predict\n",
    "from sklearn import metrics\n",
    "predicted = cross_val_predict(classifier,arr[:, idx],arr[:,1] , cv=10)\n",
    "metrics.accuracy_score(arr[:,1], predicted) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### K-Fold Cross Validation\n",
    "\n",
    "Advantages:\n",
    "- All observations are used for both training and \n",
    "validation. Each observation is used for \n",
    "validation exactly once\n",
    "\n",
    "- Non exhaustive: More tractable than leave-p-out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.99850299401197606, 0.99850074962518742, 0.992503748125937]\n"
     ]
    }
   ],
   "source": [
    "X_folds = np.array_split(Xtrain, 3)\n",
    "y_folds = np.array_split(ytrain, 3)\n",
    "scores = list()\n",
    "best_Xtrain = []\n",
    "best_ytrain = []\n",
    "best_Xtest = []\n",
    "best_ytest = []\n",
    "\n",
    "best_score=0.0\n",
    "\n",
    "for k in range(3):\n",
    "     # We use 'list' to copy, in order to 'pop' later on\n",
    "    X_train = list(X_folds)\n",
    "    X_test  = X_train.pop(k)\n",
    "    X_train = np.concatenate(X_train)\n",
    "    y_train = list(y_folds)\n",
    "    y_test  = y_train.pop(k)\n",
    "    y_train = np.concatenate(y_train)\n",
    "    score = clf.fit(X_train, y_train).score(X_test, y_test)\n",
    "    if score > best_score:\n",
    "        best_score = score\n",
    "        best_Xtrain = X_train\n",
    "        best_ytrain = y_train\n",
    "        best_Xtest = X_test\n",
    "        best_ytest = y_test\n",
    "    scores.append(score)\n",
    "    \n",
    "print(scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Best Score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.99850299401197606"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(best_Xtrain, best_ytrain).score(best_Xtest, best_ytest)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Decision Tree Classifiers\n",
    "They are very simple yet powerful supervised learning methods, which constructs a decision tree model, which will be used to make predictions.\n",
    "\n",
    "The main advantage of this model is that a human being can easily understand and reproduce the sequence of decisions (especially if the number of attributes is small) taken to predict the target class of a new instance. This is very important for tasks such as medical diagnosis or credit approval, where we want to show a reason for the decision, rather than just saying this is what the training data suggests.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cross-Validation\n",
    "\n",
    "Until now we have been splitting the dataset into a training and test set rather randomly and were reporting a rather artifical performance. Now we are going to test our system exhaustively by making use of k-fold [cross validation](http://en.wikipedia.org/wiki/Cross-validation_%28statistics%29). \n",
    "\n",
    "Now go and tune your hyper-parameters (purity, exthreshold) to opitmize the performance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('Accuracy:1.000', '\\n')\n"
     ]
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "def measure_performance(X,y,clf, show_accuracy=True, show_classification_report=True, show_confusion_matrix=True):\n",
    "    y_pred=clf.predict(X)   \n",
    "    if show_accuracy:\n",
    "        print (\"Accuracy:{0:.3f}\".format(metrics.accuracy_score(y,y_pred)))\n",
    "\n",
    "    if show_classification_report:\n",
    "        print (\"Classification report\")\n",
    "        print (metrics.classification_report(y,y_pred),\"\\n\")\n",
    "        \n",
    "    if show_confusion_matrix:\n",
    "        print (\"Confusion matrix\")\n",
    "        print (metrics.confusion_matrix(y,y_pred),\"\\n\")\n",
    "        \n",
    "measure_performance(Xtrain,ytrain,clf, show_classification_report=False, show_confusion_matrix=False)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observation: **\n",
    "\n",
    "Our tree has an accuracy of 1.0 on the training set. \n",
    "This is especially true for decision trees as this method is highly susceptible to overfitting. Since we did not separate an evaluation set, we should apply cross-validation.\n",
    "\n",
    "For this example, we will use an extreme case of crossvalidation, named leave-one-out cross-validation. For each instance in the training sample, we train on the rest of the sample, and evaluate the model built on the only instance left out. After performing as many classifications as training instances, we calculate the accuracy simply as the proportion of times our method correctly predicted the class of the left-out instance, and found it is a little lower (as we expected) than the resubstitution accuracy on the training set.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(0.7348314606741573, {'max_depth': 3})\n"
     ]
    }
   ],
   "source": [
    "# Specifying parameters for CV\n",
    "parameters = {'max_depth':range(3,20)}\n",
    "\n",
    "# We are using Grid Search to find beat paramters for beat accuracy\n",
    "clf = grid_search.GridSearchCV(tree.DecisionTreeClassifier(), parameters, n_jobs=4)\n",
    "\n",
    "#Training the model\n",
    "clf.fit(X=arr[:,idx], y=arr[:,1])\n",
    "\n",
    "#Finding the best_estimator\n",
    "tree_model = clf.best_estimator_\n",
    "\n",
    "print (clf.best_score_, clf.best_params_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### DT Algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.995515695067\n",
      "Mean Absolute Error: 0.00448430493274\n"
     ]
    }
   ],
   "source": [
    "#Creating Decision Tree Classifier\n",
    "clf = tree.DecisionTreeClassifier(max_depth=3)\n",
    "\n",
    "#Train the Model using dataset\n",
    "clf.fit(Xtrain, ytrain)\n",
    "\n",
    "#Predict the output\n",
    "yfit = clf.predict(Xtest)\n",
    "\n",
    "# Calculating Error in predictions\n",
    "mae = mean_absolute_error(ytest, yfit)\n",
    "\n",
    "print \"Accuracy = \", np.sum(yfit==ytest)/float(ytest.shape[0])\n",
    "print \"Mean Absolute Error:\",mae\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>56</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>57</td>\n",
       "      <td>33</td>\n",
       "      <td>49</td>\n",
       "      <td>50</td>\n",
       "      <td>34</td>\n",
       "      <td>223</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0   1   2   3   4  All\n",
       "Actual                            \n",
       "0          56   0   0   0   0   56\n",
       "1           1  33   0   0   0   34\n",
       "2           0   0  49   0   0   49\n",
       "3           0   0   0  50   0   50\n",
       "4           0   0   0   0  34   34\n",
       "All        57  33  49  50  34  223"
      ]
     },
     "execution_count": 155,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print_confusion_matrix(yfit.ravel(),ytest.ravel())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## K Neighbors Classifier\n",
    "\n",
    "Now we want to explore the effect of the k parameter. To do this, train the classifier multiple times, each time setting the KNN option to a different value. Try 5, 10, 50, 100, 200, 500, 1000, 1500 and 2000 and test the classifier on the test set. How does the k parameter effect the results? Hint: Consider how well the classifier is generalising to previously unseen data, and how it compares to the base rate again."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cross Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Your code goes here\n",
    "ks = [5, 10, 50, 100, 200, 500, 1000, 1500, 2000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def trainAndScoreKnnClassifier(k, trainData, trainTargets, testData, testTargets):\n",
    "    classifier = KNeighborsClassifier(n_neighbors=k)\n",
    "    classifier.fit(trainData, trainTargets)\n",
    "    #print \"knn classifier with k: \" + str(k)\n",
    "    #print \"training score:\"\n",
    "    trainingScore = classifier.score(trainData, trainTargets)\n",
    "    #print trainingScore\n",
    "    #print \"testing score:\"\n",
    "    testingScore = classifier.score(testData, testTargets)\n",
    "    #print testingScore\n",
    "    return k, trainingScore, testingScore"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "scoreDict = dict()\n",
    "for k in ks:\n",
    "    curK, curTrainingScore, curTestingScore = trainAndScoreKnnClassifier(k, Xtrain, ytrain, Xtest, ytest)\n",
    "    scoreDict[k] = {\n",
    "        \"trainingScore\": curTrainingScore,\n",
    "        \"testingScore\": curTestingScore\n",
    "    }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{5: {'testingScore': 0.99551569506726456,\n",
       "  'trainingScore': 0.99800199800199796},\n",
       " 10: {'testingScore': 0.99551569506726456,\n",
       "  'trainingScore': 0.99700299700299699},\n",
       " 50: {'testingScore': 0.99551569506726456,\n",
       "  'trainingScore': 0.9985014985014985},\n",
       " 100: {'testingScore': 0.99551569506726456,\n",
       "  'trainingScore': 0.99700299700299699},\n",
       " 200: {'testingScore': 0.99551569506726456,\n",
       "  'trainingScore': 0.99400599400599399},\n",
       " 500: {'testingScore': 0.99551569506726456,\n",
       "  'trainingScore': 0.99400599400599399},\n",
       " 1000: {'testingScore': 0.6547085201793722,\n",
       "  'trainingScore': 0.60289710289710285},\n",
       " 1500: {'testingScore': 0.47533632286995514,\n",
       "  'trainingScore': 0.45654345654345652},\n",
       " 2000: {'testingScore': 0.22421524663677131,\n",
       "  'trainingScore': 0.23026973026973027}}"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scoreDict"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "**We notice** that as the k increases the training score is decreasing, which is expected because the best prediction to be made for the training instances is the one nearest neighbor.\n",
    "\n",
    "But more importantly we care for testing, unseen, data. We see that the testing score climbs up and then starts falling again with a maximum at k=2000 even though the k=100 is much more computationally preferable and the classification accuracy is not bad.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.995515695067\n",
      "Mean Absolute Error: 0.00448430493274\n"
     ]
    }
   ],
   "source": [
    "#Creating Knn Classifier Instance\n",
    "neigh = KNeighborsClassifier(n_neighbors=200)\n",
    "\n",
    "neigh.fit(Xtrain, ytrain)\n",
    "\n",
    "\n",
    "ypredict = neigh.predict(Xtest)\n",
    "\n",
    "#Evaluating the Error in predictions\n",
    "mae =mean_absolute_error(ytest, ypredict)\n",
    "\n",
    "print \"Accuracy = \", np.sum(ypredict==ytest)/float(ytest.shape[0])\n",
    "print \"Mean Absolute Error:\",mae\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>56</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>57</td>\n",
       "      <td>33</td>\n",
       "      <td>49</td>\n",
       "      <td>50</td>\n",
       "      <td>34</td>\n",
       "      <td>223</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0   1   2   3   4  All\n",
       "Actual                            \n",
       "0          56   0   0   0   0   56\n",
       "1           1  33   0   0   0   34\n",
       "2           0   0  49   0   0   49\n",
       "3           0   0   0  50   0   50\n",
       "4           0   0   0   0  34   34\n",
       "All        57  33  49  50  34  223"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print_confusion_matrix(ypredict.ravel(),ytest.ravel())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plotting k-value Vs Accuracy\n",
    "\n",
    "Plot the results (k-value on the x-axis and classification accuracy on the y-axis), making sure to mark the axes. \n",
    "**Can you conclude anything from observing the plot?**\n",
    "**Yes,** we notice that initially when k-value is smaller, accuracy is lower but when k-value increases accuracy starts increasing untill reaching a point where with further increase in k-value leads to fall in accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAAF+CAYAAABZObp/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSZViOJEkSZViOJEkSZViOJEkSZXy/wG0BiaDuNdqIAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaae2dd090>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Your code goes here\n",
    "fig = plt.figure()\n",
    "plt.hold(True)\n",
    "plt.axis([0, 2000, 0, 1])\n",
    "plt.plot(scoreDict.keys(), [x['trainingScore'] for x in scoreDict.values()], 'b.')\n",
    "plt.plot(scoreDict.keys(), [x['testingScore'] for x in scoreDict.values()], 'r.')\n",
    "plt.xlabel('k-value')\n",
    "plt.ylabel('classification accuracy')\n",
    "plt.hold(False)\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgUAAAFoCAYAAADHHogUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSRIA/x+yO6DVk0byEgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaae212650>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting again here with log x axis to show the differences at the first small k values\n",
    "fig = plt.figure()\n",
    "plt.hold(True)\n",
    "plt.plot(np.log(scoreDict.keys()), [x['trainingScore'] for x in scoreDict.values()], 'b.')\n",
    "plt.plot(np.log(scoreDict.keys()), [x['testingScore'] for x in scoreDict.values()], 'r.')\n",
    "plt.hold(False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "Your answer goes here\n",
    "\n",
    "We notice that the training accuracy is continuously falling as the knn gets bigger while the testing accuracy rises a little bit and then falls, it has a maximum value\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Select best value for k from Questions 2.9 and 2.10 and plot the normalised confusion matrix on the test set. Then plot the confusion matrix for a 5-nearest neighbor classifier. **What do you observe?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "testingScores = [x['testingScore'] for x in scoreDict.values()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "100"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bestK = scoreDict.keys()[np.argmax(testingScores)]\n",
    "bestK"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def whatClassesAreResponsibleForEachClassification(confusionMatrix):\n",
    "    return confusionMatrix / np.sum(confusionMatrix, axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def howEachClassHasBeenClassified(confusionMatrix):\n",
    "    return (confusionMatrix.T / np.sum(confusionMatrix, axis=1)).T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Your code goes here\n",
    "def getKnnPredictions(k, trainData, trainTargets, testData):\n",
    "    classifier = KNeighborsClassifier(n_neighbors=k)\n",
    "    classifier.fit(trainData, trainTargets)\n",
    "    return classifier.predict(testData)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def getInvertedNormalizedConfusionMatrix(realTargets, predictions):\n",
    "    return whatClassesAreResponsibleForEachClassification(\n",
    "        confusion_matrix(y_true=realTargets, y_pred=predictions)\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def getNormalizedConfusionMatrix(realTargets, predictions):\n",
    "    return howEachClassHasBeenClassified(\n",
    "        confusion_matrix(y_true=realTargets, y_pred=predictions)\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(cm, classes=None, title='Confusion matrix'):\n",
    "    \"\"\"Plots a confusion matrix.\"\"\"\n",
    "    if classes is not None:\n",
    "        sns.heatmap(cm, xticklabels=classes, yticklabels=classes, vmin=0., vmax=1., annot=True)\n",
    "    else:\n",
    "        sns.heatmap(cm, vmin=0., vmax=1.)\n",
    "    plt.title(title)\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "availableClasses = np.unique(ytest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfkAAAGNCAYAAAASBQgzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaa2482ed0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "k = 1000\n",
    "\n",
    "plot_confusion_matrix(getInvertedNormalizedConfusionMatrix(\n",
    "        realTargets=ytest,\n",
    "        predictions=getKnnPredictions(k = k, trainData=Xtrain, trainTargets=ytrain, testData=Xtest)),\n",
    "    classes = availableClasses, title='Confusion Matrix for k = %d' % k)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfkAAAGNCAYAAAASBQgzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRESkpJTkRURESkpJXkREpKT+P9FqDA7dCaJeAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaa22ff1d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "k = bestK\n",
    "\n",
    "plot_confusion_matrix(getInvertedNormalizedConfusionMatrix(\n",
    "        realTargets=ytest,\n",
    "        predictions=getKnnPredictions(k = k, trainData=Xtrain, trainTargets=ytrain, testData=Xtest)),\n",
    "    classes = availableClasses, title='Confusion Matrix for k = %d' % k)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**NOTE:** Here the problem is imbalanced as we see from the histogram plotted above. The class N has large prior. We have 3 times more N instances than we do EI and IE instances. This means that as the k grows bigger the N will tend to be the majority of the instances and we are more likely to predict an instance to be of class N."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def trainAndLogLossKnnClassifier(k, trainData, trainTargets, testData, testTargets):\n",
    "    classifier = KNeighborsClassifier(n_neighbors=k)\n",
    "    classifier.fit(trainData, trainTargets)\n",
    "    #print \"knn classifier with k: \" + str(k)\n",
    "    #print \"training score:\"\n",
    "    trainingLogLoss = log_loss(y_true = trainTargets, y_pred = classifier.predict_proba(trainData))\n",
    "    #print trainingScore\n",
    "    #print \"testing score:\"\n",
    "    testingLogLoss = log_loss(y_true = testTargets, y_pred = classifier.predict_proba(testData))\n",
    "    #print testingScore\n",
    "    return k, trainingLogLoss, testingLogLoss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{5: {'testingLogLoss': 0.15688369281408118,\n",
       "  'trainingLogLoss': 0.0052536191714077996},\n",
       " 10: {'testingLogLoss': 0.16482705057004671,\n",
       "  'trainingLogLoss': 0.0078119882718811146},\n",
       " 50: {'testingLogLoss': 0.196335216833846,\n",
       "  'trainingLogLoss': 0.03079138250153704},\n",
       " 100: {'testingLogLoss': 0.23899869197804807,\n",
       "  'trainingLogLoss': 0.059988778221925665},\n",
       " 200: {'testingLogLoss': 0.30895066359652801,\n",
       "  'trainingLogLoss': 0.11950920731259643},\n",
       " 500: {'testingLogLoss': 0.38037747348100792,\n",
       "  'trainingLogLoss': 0.34505927842779821},\n",
       " 1000: {'testingLogLoss': 0.90372773675457208,\n",
       "  'trainingLogLoss': 0.90860384329536714},\n",
       " 1500: {'testingLogLoss': 1.3091928448627355,\n",
       "  'trainingLogLoss': 1.3140689514035306},\n",
       " 2000: {'testingLogLoss': 1.5968749173145151,\n",
       "  'trainingLogLoss': 1.601751023855311}}"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logLossDict = dict()\n",
    "\n",
    "for k in ks:\n",
    "    curK, curTrainingLogLoss, curTestingLogLoss = trainAndLogLossKnnClassifier(\n",
    "        k, Xtrain, ytrain, Xtest, ytest\n",
    "    )\n",
    "    \n",
    "    logLossDict[k] = {\n",
    "        \"trainingLogLoss\": curTrainingLogLoss,\n",
    "        \"testingLogLoss\": curTestingLogLoss\n",
    "    }\n",
    "    \n",
    "logLossDict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhEAAAFoCAYAAADpW6mGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaa2004390>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Your code goes here\n",
    "fig = plt.figure()\n",
    "plt.hold(True)\n",
    "plt.plot(logLossDict.keys(), [x['trainingLogLoss'] for x in logLossDict.values()], 'b.')\n",
    "plt.plot(logLossDict.keys(), [x['testingLogLoss'] for x in logLossDict.values()], 'r.')\n",
    "plt.hold(False)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgUAAAFoCAYAAADHHogUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcaa1f7c7d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Your code goes here\n",
    "fig = plt.figure()\n",
    "plt.hold(True)\n",
    "plt.plot(np.log(logLossDict.keys()), [x['trainingLogLoss'] for x in logLossDict.values()], 'b.')\n",
    "plt.plot(np.log(logLossDict.keys()), [x['testingLogLoss'] for x in logLossDict.values()], 'r.')\n",
    "plt.hold(False)\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The classification accuracy is the same for both classifiers because both would classify the one test point as correct and therefore both would have 1/1 = 100% classification accuracy.\n",
    "\n",
    "Yes indeed the classifier with probability 0.99 is better than the classifier with probability 0.55 because the classifier must be more sure that this one test point is the correct one (since it is classifying it correctly)\n",
    "\n",
    "From the plots above we notice that for training and testing data as k is small the certainty of the classifier for the testing data is quite low in comparison with the training data. While the k is getting larger the certainty (log loss) is the same for classifying training data or testing data.\n",
    "\n",
    "Now is interesting to notice that for k=50 the classifier has the minimum log loss (or the maximum certainty).\n",
    "\n",
    "In our case because the K=100 and k=500 have similar accuracy but the k=100 has lower log-loss probability we would prefer the kNN classifier with k=100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The kNearestNeighbors Classifier is a generative model, you can assign probabilities according to the neighbors.\n",
    "\n",
    "While the SVM classifier is a discriminative model which only describes the boundary between the two classes. We cannot have probabilities for it"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Linear SVC Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.757847533632\n",
      "Mean Absolute Error: 0.242152466368\n"
     ]
    }
   ],
   "source": [
    "#Creating Instance of Linear SVC Classifier\n",
    "lscv = LinearSVC()\n",
    "\n",
    "#Training the model\n",
    "lscv.fit(Xtrain, ytrain)\n",
    "\n",
    "# Testing the model\n",
    "ypredict = lscv.predict(Xtest)\n",
    "\n",
    "#Evaluating the Error in predictions\n",
    "mae =mean_absolute_error(ytest, ypredict)\n",
    "\n",
    "print \"Accuracy = \", np.sum(ypredict==ytest)/float(ytest.shape[0])\n",
    "print \"Mean Absolute Error:\",mae\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>4</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>56</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>30</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>43</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>59</td>\n",
       "      <td>30</td>\n",
       "      <td>57</td>\n",
       "      <td>77</td>\n",
       "      <td>223</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0   1   2   4  All\n",
       "Actual                        \n",
       "0          56   0   0   0   56\n",
       "1           3  30   1   0   34\n",
       "2           0   0  49   0   49\n",
       "3           0   0   7  43   50\n",
       "4           0   0   0  34   34\n",
       "All        59  30  57  77  223"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print_confusion_matrix(ypredict.ravel(),ytest.ravel())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.910313901345\n",
      "Mean Absolute Error: 0.165919282511\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "clf = RandomForestClassifier(n_estimators=10, max_depth=8)\n",
    "\n",
    "#Train the Model using dataset\n",
    "clf.fit(Xtrain, ytrain)\n",
    "\n",
    "#Predict the output\n",
    "yfit = clf.predict(Xtest)\n",
    "\n",
    "# Calculating Error in predictions\n",
    "mae = mean_absolute_error(ytest, yfit)\n",
    "\n",
    "print \"Accuracy = \", np.sum(yfit==ytest)/float(ytest.shape[0])\n",
    "print \"Mean Absolute Error:\",mae"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>55</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>23</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>46</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>29</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>60</td>\n",
       "      <td>23</td>\n",
       "      <td>47</td>\n",
       "      <td>62</td>\n",
       "      <td>31</td>\n",
       "      <td>223</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0   1   2   3   4  All\n",
       "Actual                            \n",
       "0          55   0   0   1   0   56\n",
       "1           3  23   0   6   2   34\n",
       "2           1   0  46   2   0   49\n",
       "3           0   0   0  50   0   50\n",
       "4           1   0   1   3  29   34\n",
       "All        60  23  47  62  31  223"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print_confusion_matrix(yfit.ravel(),ytest.ravel())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Gradient Boosting Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.995515695067\n",
      "Mean Absolute Error: 0.00448430493274\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "\n",
    "clf = GradientBoostingClassifier(n_estimators=100, learning_rate=1.0, max_depth=1, random_state=0)\n",
    "\n",
    "#Train the Model using dataset\n",
    "clf.fit(Xtrain, ytrain)\n",
    "\n",
    "#Predict the output\n",
    "yfit = clf.predict(Xtest)\n",
    "\n",
    "# Calculating Error in predictions\n",
    "mae = mean_absolute_error(ytest, yfit)\n",
    "\n",
    "print \"Accuracy = \", np.sum(yfit==ytest)/float(ytest.shape[0])\n",
    "print \"Mean Absolute Error:\",mae"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>56</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>57</td>\n",
       "      <td>33</td>\n",
       "      <td>49</td>\n",
       "      <td>50</td>\n",
       "      <td>34</td>\n",
       "      <td>223</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0   1   2   3   4  All\n",
       "Actual                            \n",
       "0          56   0   0   0   0   56\n",
       "1           1  33   0   0   0   34\n",
       "2           0   0  49   0   0   49\n",
       "3           0   0   0  50   0   50\n",
       "4           0   0   0   0  34   34\n",
       "All        57  33  49  50  34  223"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print_confusion_matrix(yfit.ravel(),ytest.ravel())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ada Boost Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.627802690583\n",
      "Mean Absolute Error: 0.520179372197\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "\n",
    "clf = AdaBoostClassifier()\n",
    "\n",
    "#Train the Model using dataset\n",
    "clf.fit(Xtrain, ytrain)\n",
    "\n",
    "#Predict the output\n",
    "yfit = clf.predict(Xtest)\n",
    "\n",
    "# Calculating Error in predictions\n",
    "mae = mean_absolute_error(ytest, yfit)\n",
    "\n",
    "print \"Accuracy = \", np.sum(yfit==ytest)/float(ytest.shape[0])\n",
    "print \"Mean Absolute Error:\",mae"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>56</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>57</td>\n",
       "      <td>132</td>\n",
       "      <td>34</td>\n",
       "      <td>223</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0    3   4  All\n",
       "Actual                     \n",
       "0          56    0   0   56\n",
       "1           1   33   0   34\n",
       "2           0   49   0   49\n",
       "3           0   50   0   50\n",
       "4           0    0  34   34\n",
       "All        57  132  34  223"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.print_confusion_matrix(yfit.ravel(),ytest.ravel())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Remark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "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": 1
}