activity_prediction.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from scipy.spatial.distance import cosine\n",
    "import numpy as np\n",
    "from scipy.sparse import csr_matrix\n",
    "from __future__ import division\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "# display results to 3 decimal points, not in scientific notation\n",
    "pd.set_option('display.float_format', lambda x: '%.3f' % x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "with open('/Users/nickbecker/Downloads/HAPT Data Set/features.txt') as handle:\n",
    "    features = handle.readlines()\n",
    "    features = map(lambda x: x.strip(), features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "with open('/Users/nickbecker/Downloads/HAPT Data Set/activity_labels.txt') as handle:\n",
    "    activity_labels = handle.readlines()\n",
    "    activity_labels = map(lambda x: x.strip(), activity_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "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>activity_id</th>\n",
       "      <th>activity_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>WALKING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>WALKING_UPSTAIRS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>WALKING_DOWNSTAIRS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>SITTING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>STANDING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>LAYING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>STAND_TO_SIT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>SIT_TO_STAND</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>SIT_TO_LIE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>LIE_TO_SIT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>11</td>\n",
       "      <td>STAND_TO_LIE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>12</td>\n",
       "      <td>LIE_TO_STAND</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   activity_id      activity_label\n",
       "0            1             WALKING\n",
       "1            2    WALKING_UPSTAIRS\n",
       "2            3  WALKING_DOWNSTAIRS\n",
       "3            4             SITTING\n",
       "4            5            STANDING\n",
       "5            6              LAYING\n",
       "6            7        STAND_TO_SIT\n",
       "7            8        SIT_TO_STAND\n",
       "8            9          SIT_TO_LIE\n",
       "9           10          LIE_TO_SIT\n",
       "10          11        STAND_TO_LIE\n",
       "11          12        LIE_TO_STAND"
      ]
     },
     "execution_count": 289,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "activity_df = pd.DataFrame(activity_labels)\n",
    "activity_df = pd.DataFrame(activity_df[0].str.split(' ').tolist(),\n",
    "                           columns = ['activity_id', 'activity_label'])\n",
    "activity_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "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>tBodyAcc-Mean-1</th>\n",
       "      <th>tBodyAcc-Mean-2</th>\n",
       "      <th>tBodyAcc-Mean-3</th>\n",
       "      <th>tBodyAcc-STD-1</th>\n",
       "      <th>tBodyAcc-STD-2</th>\n",
       "      <th>tBodyAcc-STD-3</th>\n",
       "      <th>tBodyAcc-Mad-1</th>\n",
       "      <th>tBodyAcc-Mad-2</th>\n",
       "      <th>tBodyAcc-Mad-3</th>\n",
       "      <th>tBodyAcc-Max-1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.044</td>\n",
       "      <td>-0.006</td>\n",
       "      <td>-0.035</td>\n",
       "      <td>-0.995</td>\n",
       "      <td>-0.988</td>\n",
       "      <td>-0.937</td>\n",
       "      <td>-0.995</td>\n",
       "      <td>-0.989</td>\n",
       "      <td>-0.953</td>\n",
       "      <td>-0.795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.039</td>\n",
       "      <td>-0.002</td>\n",
       "      <td>-0.029</td>\n",
       "      <td>-0.998</td>\n",
       "      <td>-0.983</td>\n",
       "      <td>-0.971</td>\n",
       "      <td>-0.999</td>\n",
       "      <td>-0.983</td>\n",
       "      <td>-0.974</td>\n",
       "      <td>-0.803</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.040</td>\n",
       "      <td>-0.005</td>\n",
       "      <td>-0.023</td>\n",
       "      <td>-0.995</td>\n",
       "      <td>-0.977</td>\n",
       "      <td>-0.985</td>\n",
       "      <td>-0.996</td>\n",
       "      <td>-0.976</td>\n",
       "      <td>-0.986</td>\n",
       "      <td>-0.798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.040</td>\n",
       "      <td>-0.012</td>\n",
       "      <td>-0.029</td>\n",
       "      <td>-0.996</td>\n",
       "      <td>-0.989</td>\n",
       "      <td>-0.993</td>\n",
       "      <td>-0.997</td>\n",
       "      <td>-0.989</td>\n",
       "      <td>-0.993</td>\n",
       "      <td>-0.798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.039</td>\n",
       "      <td>-0.002</td>\n",
       "      <td>-0.024</td>\n",
       "      <td>-0.998</td>\n",
       "      <td>-0.987</td>\n",
       "      <td>-0.993</td>\n",
       "      <td>-0.998</td>\n",
       "      <td>-0.986</td>\n",
       "      <td>-0.994</td>\n",
       "      <td>-0.802</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   tBodyAcc-Mean-1  tBodyAcc-Mean-2  tBodyAcc-Mean-3  tBodyAcc-STD-1  \\\n",
       "0            0.044           -0.006           -0.035          -0.995   \n",
       "1            0.039           -0.002           -0.029          -0.998   \n",
       "2            0.040           -0.005           -0.023          -0.995   \n",
       "3            0.040           -0.012           -0.029          -0.996   \n",
       "4            0.039           -0.002           -0.024          -0.998   \n",
       "\n",
       "   tBodyAcc-STD-2  tBodyAcc-STD-3  tBodyAcc-Mad-1  tBodyAcc-Mad-2  \\\n",
       "0          -0.988          -0.937          -0.995          -0.989   \n",
       "1          -0.983          -0.971          -0.999          -0.983   \n",
       "2          -0.977          -0.985          -0.996          -0.976   \n",
       "3          -0.989          -0.993          -0.997          -0.989   \n",
       "4          -0.987          -0.993          -0.998          -0.986   \n",
       "\n",
       "   tBodyAcc-Mad-3  tBodyAcc-Max-1  \n",
       "0          -0.953          -0.795  \n",
       "1          -0.974          -0.803  \n",
       "2          -0.986          -0.798  \n",
       "3          -0.993          -0.798  \n",
       "4          -0.994          -0.802  "
      ]
     },
     "execution_count": 294,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train = pd.read_table('/Users/nickbecker/Downloads/HAPT Data Set/Train/X_train.txt',\n",
    "             header = None, sep = \" \", names = features)\n",
    "x_train.iloc[:10, :10].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "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>activity_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   activity_id\n",
       "0            5\n",
       "1            5\n",
       "2            5\n",
       "3            5\n",
       "4            5"
      ]
     },
     "execution_count": 291,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train = pd.read_table('/Users/nickbecker/Downloads/HAPT Data Set/Train/y_train.txt',\n",
    "             header = None, sep = \" \", names = ['activity_id'])\n",
    "y_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "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>tBodyAcc-Mean-1</th>\n",
       "      <th>tBodyAcc-Mean-2</th>\n",
       "      <th>tBodyAcc-Mean-3</th>\n",
       "      <th>tBodyAcc-STD-1</th>\n",
       "      <th>tBodyAcc-STD-2</th>\n",
       "      <th>tBodyAcc-STD-3</th>\n",
       "      <th>tBodyAcc-Mad-1</th>\n",
       "      <th>tBodyAcc-Mad-2</th>\n",
       "      <th>tBodyAcc-Mad-3</th>\n",
       "      <th>tBodyAcc-Max-1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.031</td>\n",
       "      <td>-0.009</td>\n",
       "      <td>0.040</td>\n",
       "      <td>-0.939</td>\n",
       "      <td>-0.945</td>\n",
       "      <td>-0.759</td>\n",
       "      <td>-0.952</td>\n",
       "      <td>-0.950</td>\n",
       "      <td>-0.802</td>\n",
       "      <td>-0.757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.043</td>\n",
       "      <td>0.001</td>\n",
       "      <td>-0.026</td>\n",
       "      <td>-0.976</td>\n",
       "      <td>-0.978</td>\n",
       "      <td>-0.960</td>\n",
       "      <td>-0.987</td>\n",
       "      <td>-0.979</td>\n",
       "      <td>-0.967</td>\n",
       "      <td>-0.757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.038</td>\n",
       "      <td>-0.012</td>\n",
       "      <td>-0.026</td>\n",
       "      <td>-0.994</td>\n",
       "      <td>-0.979</td>\n",
       "      <td>-0.973</td>\n",
       "      <td>-0.994</td>\n",
       "      <td>-0.981</td>\n",
       "      <td>-0.978</td>\n",
       "      <td>-0.799</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.036</td>\n",
       "      <td>-0.018</td>\n",
       "      <td>-0.025</td>\n",
       "      <td>-0.995</td>\n",
       "      <td>-0.982</td>\n",
       "      <td>-0.976</td>\n",
       "      <td>-0.995</td>\n",
       "      <td>-0.983</td>\n",
       "      <td>-0.981</td>\n",
       "      <td>-0.798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.038</td>\n",
       "      <td>-0.013</td>\n",
       "      <td>-0.033</td>\n",
       "      <td>-0.994</td>\n",
       "      <td>-0.977</td>\n",
       "      <td>-0.984</td>\n",
       "      <td>-0.994</td>\n",
       "      <td>-0.977</td>\n",
       "      <td>-0.986</td>\n",
       "      <td>-0.798</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   tBodyAcc-Mean-1  tBodyAcc-Mean-2  tBodyAcc-Mean-3  tBodyAcc-STD-1  \\\n",
       "0            0.031           -0.009            0.040          -0.939   \n",
       "1            0.043            0.001           -0.026          -0.976   \n",
       "2            0.038           -0.012           -0.026          -0.994   \n",
       "3            0.036           -0.018           -0.025          -0.995   \n",
       "4            0.038           -0.013           -0.033          -0.994   \n",
       "\n",
       "   tBodyAcc-STD-2  tBodyAcc-STD-3  tBodyAcc-Mad-1  tBodyAcc-Mad-2  \\\n",
       "0          -0.945          -0.759          -0.952          -0.950   \n",
       "1          -0.978          -0.960          -0.987          -0.979   \n",
       "2          -0.979          -0.973          -0.994          -0.981   \n",
       "3          -0.982          -0.976          -0.995          -0.983   \n",
       "4          -0.977          -0.984          -0.994          -0.977   \n",
       "\n",
       "   tBodyAcc-Mad-3  tBodyAcc-Max-1  \n",
       "0          -0.802          -0.757  \n",
       "1          -0.967          -0.757  \n",
       "2          -0.978          -0.799  \n",
       "3          -0.981          -0.798  \n",
       "4          -0.986          -0.798  "
      ]
     },
     "execution_count": 295,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_test = pd.read_table('/Users/nickbecker/Downloads/HAPT Data Set/Test/X_test.txt',\n",
    "             header = None, sep = \" \", names = features)\n",
    "x_test.iloc[:10, :10].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "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>activity_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   activity_id\n",
       "0            5\n",
       "1            5\n",
       "2            5\n",
       "3            5\n",
       "4            5"
      ]
     },
     "execution_count": 242,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test = pd.read_table('/Users/nickbecker/Downloads/HAPT Data Set/Test/y_test.txt',\n",
    "             header = None, sep = \" \", names = ['activity_id'])\n",
    "y_test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Building a Human Activity Classifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we've got the train and test data loaded into memory, we can start building a model to predict the activity from the acceleration and angular velocity features.\n",
    "\n",
    "Let's build a model and plot the cross-validation accuracy curves for the training data. We'll split `x_train` and `y_train` into training and validation sets since we only want to predict on the test set once. We'll use 5-fold cross validation to get a good sense of the model's accuracy at different values of C. To plot the curves, we'll use the `validation_curve` function from sklearn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "from sklearn.learning_curve import validation_curve\n",
    "\n",
    "C_params = np.logspace(-6, 3, 10)\n",
    "svc_2 = LinearSVC(random_state = 12)\n",
    "\n",
    "train_scores, test_scores = validation_curve(\n",
    "    svc_2, x_train.values, y_train.values.flatten(),\n",
    "    param_name=\"C\", param_range=C_params,\n",
    "    cv=5, scoring=\"accuracy\", n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train_scores_mean = np.mean(train_scores, axis=1)\n",
    "train_scores_std = np.std(train_scores, axis=1)\n",
    "test_scores_mean = np.mean(test_scores, axis=1)\n",
    "test_scores_std = np.std(test_scores, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 322,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 341,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAukAAAIICAYAAADT3HAsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x130511f10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_min = 0.5\n",
    "y_max = 1.0\n",
    "\n",
    "f = plt.figure(figsize = (12, 8))\n",
    "ax = plt.axes()\n",
    "sns.set(font_scale = 1.25)\n",
    "sns.set_style(\"darkgrid\")\n",
    "plt.title(\"SVM Training and Validation Accuracy\")\n",
    "plt.xlabel(\"C Value\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.ylim(y_min, y_max)\n",
    "plt.yticks(np.arange(y_min, y_max + .01, .05))\n",
    "plt.semilogx(C_params, train_scores_mean, label=\"CV Training Accuracy\", color=\"red\")\n",
    "plt.fill_between(C_params, train_scores_mean - train_scores_std,\n",
    "                 train_scores_mean + train_scores_std, alpha=0.2, color=\"red\")\n",
    "plt.semilogx(C_params, test_scores_mean, label=\"CV Validation Accuracy\",\n",
    "             color=\"green\")\n",
    "plt.fill_between(C_params, test_scores_mean - test_scores_std,\n",
    "                 test_scores_mean + test_scores_std, alpha=0.2, color=\"green\")\n",
    "plt.legend(loc=\"best\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the graph, it looks like the best value of C is at 10<sup>-1</sup>. The validation accuracy begins slowly decreasing after that 10<sup>-1</sup>, indicating we are overfitting. The validation curve looks great, but we only optimized on C. We don't have to use a linear kernel. We could do a grid search on different kernels and C values. With a larger search space, we might get a different set of optimal parameters.\n",
    "\n",
    "Let's run a parameter grid search to see what the optimal parameters are for the SVM model. We'll switch to the general Support Vector Classifier in `sklearn` so we can have the option to try non-linear kernels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise',\n",
       "       estimator=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape=None, degree=3, gamma='auto', kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=12, shrinking=True,\n",
       "  tol=0.001, verbose=False),\n",
       "       fit_params={}, iid=True, n_jobs=-1,\n",
       "       param_grid=[{'kernel': ['rbf'], 'C': array([  1.00000e-06,   1.00000e-05,   1.00000e-04,   1.00000e-03,\n",
       "         1.00000e-02,   1.00000e-01,   1.00000e+00,   1.00000e+01,\n",
       "         1.00000e+02,   1.00000e+03])}, {'kernel': ['linear'], 'C': array([  1.00000e-06,   1.00000e-05,   1.00000e-04,   1.00000e-03,\n",
       "         1.00000e-02,   1.00000e-01,   1.00000e+00,   1.00000e+01,\n",
       "         1.00000e+02,   1.00000e+03])}],\n",
       "       pre_dispatch='2*n_jobs', refit=True, scoring=None, verbose=0)"
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "\n",
    "Cs = np.logspace(-6, 3, 10)\n",
    "parameters = [{'kernel': ['rbf'], 'C': Cs},\n",
    "              {'kernel': ['linear'], 'C': Cs}]\n",
    "\n",
    "svc = SVC(random_state = 12)\n",
    "\n",
    "clf = GridSearchCV(estimator = svc, param_grid = parameters, cv = 5, n_jobs = -1)\n",
    "clf.fit(x_train.values, y_train.values.flatten())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVC(C=1000.0, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma='auto', kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=12, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "{'kernel': 'rbf', 'C': 1000.0}\n",
      "0.925968842539\n"
     ]
    }
   ],
   "source": [
    "print clf.best_estimator_\n",
    "print clf.best_params_\n",
    "print clf.best_score_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So the best cross-validated parameters are a `rbf` kernel with `C = 1000`. Let's use the best model to predict on the test data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.94339025932953824"
      ]
     },
     "execution_count": 149,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.score(x_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "94% accuracy. That seems pretty high. But how do know how good that actually is?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluating the Model\n",
    "\n",
    "How much better is it than the no-information rate? The no-information rate is the accuracy we'd get if we guessed the most common class for every observation. It's the best we could do with no features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.17583807716635041"
      ]
     },
     "execution_count": 182,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.activity_id.value_counts().values[0] / y_test.activity_id.value_counts().values.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our model's accuracy of 94% looks even better compared to the baseline of 18%. That's over five times more accurate than the no-information accuracy. But where does our model fail? Let's look at the crosstab."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "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>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>1</th>\n",
       "      <td>488</td>\n",
       "      <td>4</td>\n",
       "      <td>4</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>496</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>20</td>\n",
       "      <td>449</td>\n",
       "      <td>1</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>0</td>\n",
       "      <td>471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>16</td>\n",
       "      <td>399</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>420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>453</td>\n",
       "      <td>52</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>508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>19</td>\n",
       "      <td>536</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>556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>544</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>545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\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>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\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>19</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\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>9</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>513</td>\n",
       "      <td>472</td>\n",
       "      <td>404</td>\n",
       "      <td>476</td>\n",
       "      <td>588</td>\n",
       "      <td>545</td>\n",
       "      <td>20</td>\n",
       "      <td>12</td>\n",
       "      <td>32</td>\n",
       "      <td>29</td>\n",
       "      <td>50</td>\n",
       "      <td>21</td>\n",
       "      <td>3162</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted    1    2    3    4    5    6   7   8   9  10  11  12   All\n",
       "True                                                                 \n",
       "1          488    4    4    0    0    0   0   0   0   0   0   0   496\n",
       "2           20  449    1    0    0    0   1   0   0   0   0   0   471\n",
       "3            5   16  399    0    0    0   0   0   0   0   0   0   420\n",
       "4            0    2    0  453   52    0   0   1   0   0   0   0   508\n",
       "5            0    0    0   19  536    0   1   0   0   0   0   0   556\n",
       "6            0    0    0    0    0  544   0   0   0   0   1   0   545\n",
       "7            0    1    0    3    0    0  16   1   1   0   1   0    23\n",
       "8            0    0    0    0    0    0   0  10   0   0   0   0    10\n",
       "9            0    0    0    0    0    0   1   0  20   0  11   0    32\n",
       "10           0    0    0    0    0    0   0   0   0  19   0   6    25\n",
       "11           0    0    0    1    0    1   1   0  11   1  34   0    49\n",
       "12           0    0    0    0    0    0   0   0   0   9   3  15    27\n",
       "All        513  472  404  476  588  545  20  12  32  29  50  21  3162"
      ]
     },
     "execution_count": 192,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crosstab = pd.crosstab(y_test.values.flatten(), clf.predict(x_test),\n",
    "                          rownames=['True'], colnames=['Predicted'],\n",
    "                          margins=True)\n",
    "crosstab"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We do really well for `activity_ids` 1-3, 5-6, and 8, but much worse for 4, 7 and 9-12. Why is that?\n",
    "\n",
    "One possible answer is that we don't have enough data. The `All` column on the right side of the crosstab bears that out. We have way fewer observations for activities 7-12.\n",
    "\n",
    "Additionally, it's clear the model seems to be systematically mistaking some activities for others (activities 4 and 5, 9 and 11, and 10 and 12 are confused for each other more than others). Maybe there's a reason for this. Let's put the labels on the `activity_ids` and see if we notice any patterns.\n",
    "\n",
    "I'll also convert the crosstab to percentages to make it a bit easier to interpret."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "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>WALKING</th>\n",
       "      <th>WALKING_UPSTAIRS</th>\n",
       "      <th>WALKING_DOWNSTAIRS</th>\n",
       "      <th>SITTING</th>\n",
       "      <th>STANDING</th>\n",
       "      <th>LAYING</th>\n",
       "      <th>STAND_TO_SIT</th>\n",
       "      <th>SIT_TO_STAND</th>\n",
       "      <th>SIT_TO_LIE</th>\n",
       "      <th>LIE_TO_SIT</th>\n",
       "      <th>STAND_TO_LIE</th>\n",
       "      <th>LIE_TO_STAND</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>WALKING</th>\n",
       "      <td>488</td>\n",
       "      <td>4</td>\n",
       "      <td>4</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>WALKING_UPSTAIRS</th>\n",
       "      <td>20</td>\n",
       "      <td>449</td>\n",
       "      <td>1</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>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WALKING_DOWNSTAIRS</th>\n",
       "      <td>5</td>\n",
       "      <td>16</td>\n",
       "      <td>399</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>SITTING</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>453</td>\n",
       "      <td>52</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STANDING</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>19</td>\n",
       "      <td>536</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAYING</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>544</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STAND_TO_SIT</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SIT_TO_STAND</th>\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>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SIT_TO_LIE</th>\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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LIE_TO_SIT</th>\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>19</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STAND_TO_LIE</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LIE_TO_STAND</th>\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>9</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    WALKING  WALKING_UPSTAIRS  WALKING_DOWNSTAIRS  SITTING  \\\n",
       "WALKING                 488                 4                   4        0   \n",
       "WALKING_UPSTAIRS         20               449                   1        0   \n",
       "WALKING_DOWNSTAIRS        5                16                 399        0   \n",
       "SITTING                   0                 2                   0      453   \n",
       "STANDING                  0                 0                   0       19   \n",
       "LAYING                    0                 0                   0        0   \n",
       "STAND_TO_SIT              0                 1                   0        3   \n",
       "SIT_TO_STAND              0                 0                   0        0   \n",
       "SIT_TO_LIE                0                 0                   0        0   \n",
       "LIE_TO_SIT                0                 0                   0        0   \n",
       "STAND_TO_LIE              0                 0                   0        1   \n",
       "LIE_TO_STAND              0                 0                   0        0   \n",
       "\n",
       "                    STANDING  LAYING  STAND_TO_SIT  SIT_TO_STAND  SIT_TO_LIE  \\\n",
       "WALKING                    0       0             0             0           0   \n",
       "WALKING_UPSTAIRS           0       0             1             0           0   \n",
       "WALKING_DOWNSTAIRS         0       0             0             0           0   \n",
       "SITTING                   52       0             0             1           0   \n",
       "STANDING                 536       0             1             0           0   \n",
       "LAYING                     0     544             0             0           0   \n",
       "STAND_TO_SIT               0       0            16             1           1   \n",
       "SIT_TO_STAND               0       0             0            10           0   \n",
       "SIT_TO_LIE                 0       0             1             0          20   \n",
       "LIE_TO_SIT                 0       0             0             0           0   \n",
       "STAND_TO_LIE               0       1             1             0          11   \n",
       "LIE_TO_STAND               0       0             0             0           0   \n",
       "\n",
       "                    LIE_TO_SIT  STAND_TO_LIE  LIE_TO_STAND  \n",
       "WALKING                      0             0             0  \n",
       "WALKING_UPSTAIRS             0             0             0  \n",
       "WALKING_DOWNSTAIRS           0             0             0  \n",
       "SITTING                      0             0             0  \n",
       "STANDING                     0             0             0  \n",
       "LAYING                       0             1             0  \n",
       "STAND_TO_SIT                 0             1             0  \n",
       "SIT_TO_STAND                 0             0             0  \n",
       "SIT_TO_LIE                   0            11             0  \n",
       "LIE_TO_SIT                  19             0             6  \n",
       "STAND_TO_LIE                 1            34             0  \n",
       "LIE_TO_STAND                 9             3            15  "
      ]
     },
     "execution_count": 280,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#percentages_crosstab = crosstab.iloc[:-1, :-1].apply(lambda x: x/x.sum(), axis = 1)\n",
    "percentages_crosstab = crosstab.iloc[:-1, :-1]\n",
    "percentages_crosstab.columns = activity_df.activity_label.values\n",
    "percentages_crosstab.index = activity_df.activity_label.values\n",
    "percentages_crosstab"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That makes it way more clear. The model struggles to classify those `activity_ids` because they are similar actions. 4 and 5 are both stationary (sitting and standing), 9 and 11 both involving lying down (sit-to-lie and stand-to-lie), and 10 and 12 both involve standing up from a resting position (lie-to-sit and lie_to_stand). It makes sense that accelerometer data on these actions would be similar."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So with 94% accuracy in this activity classifier scenario, can we start a Fitbit competitor? Maybe.\n",
    "\n",
    "We don't necessarily need to distinguish between the 12 different activities themselves, only whether we should count it as walking. If we can predict whether someone is walking or not from the accelerometer with near perfect accuracy, we'd be in business."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predicting Walking vs. Not Walking\n",
    "\n",
    "So how do we classify walking? First we need to convert the labels to a binary to indicate whether they represent walking or staying in place. From the activity labels, we know that 1-3 are walking and everything other activity involves staying in place."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "y_train['walking_flag'] = y_train.activity_id.apply(lambda x: 1 if x <= 3 else 0)\n",
    "y_test['walking_flag'] = y_test.activity_id.apply(lambda x: 1 if x <= 3 else 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, we'll train a new SVM model and find the optimal parameters with cross-validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n",
      "/Users/nickbecker/anaconda/lib/python2.7/site-packages/sklearn/externals/joblib/hashing.py:197: DeprecationWarning: Changing the shape of non-C contiguous array by\n",
      "descriptor assignment is deprecated. To maintain\n",
      "the Fortran contiguity of a multidimensional Fortran\n",
      "array, use 'a.T.view(...).T' instead\n",
      "  obj_bytes_view = obj.view(self.np.uint8)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise',\n",
       "       estimator=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape=None, degree=3, gamma='auto', kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=12, shrinking=True,\n",
       "  tol=0.001, verbose=False),\n",
       "       fit_params={}, iid=True, n_jobs=-1,\n",
       "       param_grid=[{'kernel': ['rbf'], 'C': array([  1.00000e-06,   1.00000e-05,   1.00000e-04,   1.00000e-03,\n",
       "         1.00000e-02,   1.00000e-01,   1.00000e+00,   1.00000e+01,\n",
       "         1.00000e+02,   1.00000e+03])}, {'kernel': ['linear'], 'C': array([  1.00000e-06,   1.00000e-05,   1.00000e-04,   1.00000e-03,\n",
       "         1.00000e-02,   1.00000e-01,   1.00000e+00,   1.00000e+01,\n",
       "         1.00000e+02,   1.00000e+03])}],\n",
       "       pre_dispatch='2*n_jobs', refit=True, scoring=None, verbose=0)"
      ]
     },
     "execution_count": 248,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "\n",
    "Cs = np.logspace(-6, 3, 10)\n",
    "parameters = [{'kernel': ['rbf'], 'C': Cs},\n",
    "              {'kernel': ['linear'], 'C': Cs}]\n",
    "\n",
    "svc = SVC(random_state = 12)\n",
    "\n",
    "clf_binary = GridSearchCV(estimator = svc, param_grid = parameters, cv = 5, n_jobs = -1)\n",
    "clf_binary.fit(x_train.values, y_train.walking_flag.values.flatten())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVC(C=0.10000000000000001, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma='auto', kernel='linear',\n",
      "  max_iter=-1, probability=False, random_state=12, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "{'kernel': 'linear', 'C': 0.10000000000000001}\n",
      "0.999098751127\n"
     ]
    }
   ],
   "source": [
    "print clf_binary.best_estimator_\n",
    "print clf_binary.best_params_\n",
    "print clf_binary.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.99905123339658441"
      ]
     },
     "execution_count": 251,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_binary.score(x_test, y_test.walking_flag)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "99.9% accuracy! Let's look at the crosstab."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "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>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1774</td>\n",
       "      <td>1</td>\n",
       "      <td>1775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1385</td>\n",
       "      <td>1387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>1776</td>\n",
       "      <td>1386</td>\n",
       "      <td>3162</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted     0     1   All\n",
       "True                       \n",
       "0          1774     1  1775\n",
       "1             2  1385  1387\n",
       "All        1776  1386  3162"
      ]
     },
     "execution_count": 253,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crosstab_binary = pd.crosstab(y_test.walking_flag.values.flatten(), clf_binary.predict(x_test),\n",
    "                          rownames=['True'], colnames=['Predicted'],\n",
    "                          margins=True)\n",
    "crosstab_binary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That's beautiful! We can almost perfectly distinguish walking from staying in place from smartphone accelerometer data. Now we just need to measure the distance traveled and we're ready to compete with Fitbit."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [Root]",
   "language": "python",
   "name": "Python [Root]"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}