Clustering.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os\n",
    "\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib inline\n",
    "plt.rcParams['figure.figsize'] = (8, 8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "orderdf = pd.read_csv('ignored_assets/orders.csv', parse_dates = [6,7])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "counts = orderdf[['driver_id', 'order_id']][~pd.isnull(orderdf['driver_id'])].groupby(['driver_id']).count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1032d2f50>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEQCAYAAABIqvhxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x107d8c190>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "orderdf[orderdf['driver_id'] == '0003191f87382a164e4d8e0378e9ef05'][['price', 'timestamp']].set_index('timestamp').plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "driver1 = orderdf[orderdf['driver_id'] == '0003191f87382a164e4d8e0378e9ef05']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ramoniglesias/development/cs341/venv/lib/python2.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n",
      "/Users/ramoniglesias/development/cs341/venv/lib/python2.7/site-packages/ipykernel_launcher.py:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "driver1['hour'] = driver1['timestamp'].dt.hour\n",
    "driver1['dayofweek'] = driver1['timestamp'].dt.dayofweek"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "ptable = pd.pivot_table(driver1, index = ['hour'], columns=['dayofweek'], values = ['order_id'],aggfunc=len)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"7\" halign=\"left\">order_id</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dayofweek</th>\n",
       "      <th>0</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hour</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>0</th>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>5.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>NaN</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          order_id                              \n",
       "dayofweek        0    1    2    3    4    5    6\n",
       "hour                                            \n",
       "0              3.0  NaN  5.0  6.0  4.0  2.0  2.0\n",
       "1              4.0  NaN  5.0  5.0  3.0  NaN  2.0\n",
       "2              1.0  NaN  4.0  3.0  NaN  NaN  3.0\n",
       "3              NaN  NaN  1.0  3.0  NaN  NaN  NaN\n",
       "11             5.0  1.0  NaN  NaN  NaN  NaN  NaN\n",
       "12             1.0  NaN  NaN  NaN  NaN  NaN  NaN\n",
       "13             NaN  NaN  1.0  NaN  NaN  NaN  NaN\n",
       "14             2.0  NaN  NaN  NaN  NaN  NaN  NaN\n",
       "15             2.0  NaN  NaN  1.0  NaN  2.0  NaN\n",
       "16             1.0  NaN  1.0  NaN  NaN  1.0  NaN\n",
       "17             NaN  NaN  1.0  3.0  NaN  NaN  2.0\n",
       "18             2.0  NaN  NaN  NaN  NaN  2.0  3.0\n",
       "19             NaN  3.0  2.0  2.0  NaN  NaN  5.0\n",
       "20             NaN  1.0  2.0  1.0  NaN  NaN  2.0\n",
       "21             1.0  NaN  2.0  1.0  2.0  NaN  NaN\n",
       "22             3.0  3.0  1.0  1.0  2.0  NaN  2.0\n",
       "23             NaN  5.0  2.0  3.0  4.0  5.0  4.0"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ptable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x179674c50>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAH8AAAD8CAYAAABemXtlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10f7a3810>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(ptable)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "orderdf['hour'] = orderdf['timestamp'].dt.hour\n",
    "orderdf['dayofweek'] = orderdf['timestamp'].dt.dayofweek"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "bigtable = pd.pivot_table(orderdf, index = ['hour'], columns=['driver_id'], values = ['order_id'],aggfunc=len)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sums = bigtable.fillna(0).as_matrix().sum(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'sort_indices' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-13-e1f5c7b3f43d>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcumsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msums\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0msort_indices\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'sort_indices' is not defined"
     ]
    }
   ],
   "source": [
    "plt.plot(np.cumsum(sums[sort_indices]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAD8CAYAAACcjGjIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nOoaDJKljOEiSOoaDJKnz/4nC9RVZQN+MAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1640e9990>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(sums, bins = 100);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sort_indices = np.argsort(sums)[::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x157227110>"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAACOCAYAAADgtntxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGqALsqIoiqKkAf8f6Kuj0d9THMIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x16a7ab2d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(bigtable.as_matrix()[:,sort_indices[:100]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ nan,  nan,  nan,  nan,  nan,  nan,  nan,  53.,  68.,  63.,  49.,\n",
       "        34.,  59.,  47.,  36.,  34.,  69.,  66.,  66.,  39.,  42.,  29.,\n",
       "        15.,  nan])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bigtable.as_matrix()[:,sort_indices[10]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "early_sums = bigtable.fillna(0).as_matrix()[:6,:].sum(axis=0)\n",
    "early_birds = np.argsort(early_sums)[::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x161350f90>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAABzCAYAAACIEflfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMAyjSrAB3TAMo0r4/6+ec+bMJarvAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x118010310>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(bigtable.as_matrix()[:,early_birds[:100]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Clustering based on hour of operation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's cluster them based on when they prefer to work. There are two options, to include _also_ how often they work, or to simply regularize their frequency. Let's start by using both."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KMeans by hour, unnormalized"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sample = bigtable.fillna(0).as_matrix().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "nclusters = 5\n",
    "kmeans = KMeans(n_clusters=nclusters, random_state=0).fit(sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1207c9d10>"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAHVCAYAAAAgiIjxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1466eee10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(nclusters):\n",
    "    plt.plot(kmeans.cluster_centers_[i], label=str(i) + ', ' + str(np.sum(kmeans.labels_ == i)) + ' of drivers')\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, it seems that it did catch some patterns, let's see how consistent are the distributions. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x18b59f2d0>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD8CAYAAACb4nSYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nzcwS4tA3M0uIQ9/MLCEOfTOzhPw/mizMkmO8RjgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11e728450>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(nclusters):\n",
    "    if i in [1]: continue\n",
    "        \n",
    "    distr = sample[kmeans.labels_ == i].sum(axis=0) / sample[kmeans.labels_ == i].sum(axis=0).sum()\n",
    "    plt.bar(range(24),distr, alpha=0.2, label=str(i))\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KMeans by hour, normalized"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# normalized and smoothed\n",
    "s = 5\n",
    "norm_sample = (sample + s) / (sample.sum(axis=1)[:,None] + s*24)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "nclusters = 5\n",
    "norm_kmeans = KMeans(n_clusters=nclusters, random_state=0).fit(norm_sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x149c601d0>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD8CAYAAACb4nSYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nq6ioqLQi/h+5v6TILlr1eQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x17c6e0350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(nclusters):\n",
    "    plt.plot(norm_kmeans.cluster_centers_[i], label=str(i) + ', ' + str(np.sum(norm_kmeans.labels_ == i)) + ' drivers')\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x17841c550>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD8CAYAAACb4nSYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n6IuIhIiCvohIiCjoi4iEiIK+iEiI/H9Xy3KNoJIlpwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x183b06210>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(nclusters):\n",
    "    if i ==0: continue\n",
    "        \n",
    "    distr = sample[norm_kmeans.labels_ == i].sum(axis=0) / sample[norm_kmeans.labels_ == i].sum(axis=0).sum()\n",
    "    plt.bar(range(24),distr, alpha=0.2, label=str(i) + ', ' + str(np.sum(norm_kmeans.labels_ == i)) + ' drivers')\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The normalized and smoothed version works much better. The smoothing helps catching the lousy drivers and that way we can discarded. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 360,
   "metadata": {},
   "outputs": [
    {
     "ename": "IndexError",
     "evalue": "index 460359 is out of bounds for axis 1 with size 93089",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-360-fdfb26b02263>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mcumdistr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcumsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msums\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0msort_indices\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0msums\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcumdistr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0msorted_labels\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnorm_kmeans\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlabels_\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0msort_indices\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscatter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margwhere\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msorted_labels\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mcumdistr\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0msorted_labels\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmarker\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'x'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcolor\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'green'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0malpha\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0.5\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mIndexError\u001b[0m: index 460359 is out of bounds for axis 1 with size 93089"
     ]
    }
   ],
   "source": [
    "cumdistr = np.cumsum(sums[sort_indices]) / sums.sum()\n",
    "plt.plot(cumdistr)\n",
    "sorted_labels = norm_kmeans.labels_[sort_indices]\n",
    "plt.scatter(np.argwhere(sorted_labels == 0),cumdistr[sorted_labels == 0], marker='x', color='green', alpha=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 316,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x34c337a90>"
      ]
     },
     "execution_count": 316,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x34afa78d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cumdistr = np.cumsum(sums[sort_indices]) / sums.sum()\n",
    "plt.plot(cumdistr)\n",
    "sorted_labels = kmeans.labels_[sort_indices]\n",
    "plt.scatter(np.argwhere(sorted_labels == 1),cumdistr[sorted_labels == 1], marker='x', color='green', alpha=0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While the smoothed version is able to better capture specific behaviors, it takes from the entire set, and can only distinguish the behaviors from those with sufficient rides. The unnormalized is able to capture both behavior and the different "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "match = {did[1]: kmeans.labels_[i] for i,did in enumerate(bigtable.columns)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def assign_driver_label(dids):\n",
    "    labels = np.zeros(len(dids), dtype=object)\n",
    "    for i,did in enumerate(dids):\n",
    "        if pd.isnull(did):\n",
    "            labels[i] = np.nan\n",
    "        else:\n",
    "            labels[i] = match[did]\n",
    "    return labels\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "orderdf['driver_cluster'] = assign_driver_label(orderdf['driver_id'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "orderdf.to_csv('ignored_assets/orders.csv', index = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Rider Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 320,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "paxtable = pd.pivot_table(orderdf, index = ['hour'], columns=['passenger_id'], values = ['order_id'],aggfunc=len)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 324,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "paxsums = paxtable.fillna(0).as_matrix().sum(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 325,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "paxsort_indices = np.argsort(sums)[::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 328,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x339f1db90>"
      ]
     },
     "execution_count": 328,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAB0CAYAAACVFMnnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nmHXNDS7gAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x34c530c10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(paxtable.as_matrix()[:,sort_indices[1:100]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 329,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "paxearly_sums = paxtable.fillna(0).as_matrix()[:6,:].sum(axis=0)\n",
    "paxearly_birds = np.argsort(paxearly_sums)[::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 331,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x33a106dd0>"
      ]
     },
     "execution_count": 331,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAB0CAYAAACVFMnnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCIKQEKRDFwRBSAjSoQuCICSE/wf5tPDg1UT/6wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x339f8cb10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(paxtable.as_matrix()[:,paxearly_birds[1:100]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Kmeans passenger, unnormalized"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 332,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "paxsample = paxtable.fillna(0).as_matrix().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 346,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "nclusters = 4\n",
    "pax_kmeans = KMeans(n_clusters=nclusters, random_state=0).fit(paxsample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 359,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x36f8147d0>"
      ]
     },
     "execution_count": 359,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x3385f2c10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(nclusters):\n",
    "    if i == 1: continue\n",
    "    plt.plot(pax_kmeans.cluster_centers_[i], label=str(i) + ', ' + str(np.sum(pax_kmeans.labels_ == i)) + ' riders')\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 357,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x36d6c0510>"
      ]
     },
     "execution_count": 357,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAEF5JREFUeJzt3W2MHeV5h/Hrjl+wFDtJazYR8ZrYBKtgF2SMDXxwaaji\nxnYjO8WEGKUNVUBuJSy5Iq1qtRVNqBo1qUipBKpC2igpamrR9AVLdU2tkKptpAQvsRUw1MIBJz4O\nNcYJCSgY7OXuhz0mx4uXnd2dc86e51y/Lzsvz87cY8/+d84zM89GZiJJKstbul2AJKl+hrskFchw\nl6QCGe6SVCDDXZIKZLhLUoEMd0kqkOEuSQUy3CWpQDO7tePzzz8/Fy1a1K3dS1JPevTRR5/PzIHx\n2nUt3BctWsTQ0FC3di9JPSkivlelnd0yklQgw12SCmS4S1KButbnLkndcurUKRqNBidPnux2KWOa\nM2cOg4ODzJo1a1Lfb7hL6juNRoN58+axaNEiIqLb5bxBZnLixAkajQaLFy+e1DbslpHUd06ePMn8\n+fOnZbADRATz58+f0icLw11SX5quwX7GVOsz3CWpQPa5S+p7jzV+XOv2Lht8+7htdu/ezbZt2xge\nHubWW29l+/bttdZguKtWVX9Iqpz8UqmGh4e57bbb2LNnD4ODg6xatYoNGzawdOnS2vZht4wkddgj\njzzCxRdfzEUXXcTs2bPZvHkzDz74YK37MNwlqcOOHj3KwoULX58fHBzk6NGjte7DcJekAhnuktRh\nCxYs4MiRI6/PNxoNFixYUOs+DHdJ6rBVq1bx1FNP8cwzz/Dqq6+yY8cONmzYUOs+fFpGUt/r9NNb\nM2fO5J577uEDH/gAw8PDfPzjH2fZsmX17qPWrUmSKlm/fj3r169v2/btlpGkAhnuklQgw12SCmS4\nS1KBDHdJKpDhLkkF8lFISfrBvnq39+4r3nT1kSNH+NjHPsaxY8eICLZs2cK2bdtqLcFwl6QOmzlz\nJnfddRcrVqzgxRdf5Morr2TNmjUO+StJveyCCy5gxYoVAMybN49LL73UUSElqSSHDx9m3759XH31\n1bVu13CXpC556aWX2LRpE3fffTdve9vbat12pXCPiLURcTAiDkXEmH/oLyI2RURGxMr6SpSk8pw6\ndYpNmzbx0Y9+lOuvv7727Y8b7hExA7gXWAcsBW6KiDf0+kfEPGAb8K26i5SkkmQmt9xyC5deeim3\n3357W/ZR5WmZq4BDmfk0QETsADYCT4xq96fAZ4Dfr7VCSWq3cR5drNs3vvEN7r//fi677DKWL18O\nwKc//elaR4msEu4LgCMt8w3grJ7/iFgBLMzMf4sIw12S3sTq1avJzLbuY8o3VCPiLcDngE9UaLsl\nIoYiYuj48eNT3bUkaQxVwv0osLBlfrC57Ix5wC8C/xkRh4FrgJ3nuqmamfdl5srMXDkwMDD5qiVJ\nb6pKuO8FlkTE4oiYDWwGdp5ZmZk/zszzM3NRZi4CvglsyMyhtlQsSRrXuOGemaeBrcBDwJPAA5l5\nICLujIh6/6KrJKkWlcaWycxdwK5Ry+4Yo+37pl6WJGkqfENVkgrkqJCS+t6BEwdq3d6y+cvedP3J\nkye59tpreeWVVzh9+jQ33HADn/rUp2qtwXCXpA4777zzePjhh5k7dy6nTp1i9erVrFu3jmuuuaa2\nfdgtI0kdFhHMnTsXGBlj5tSpU0RErfsw3CWpC4aHh1m+fDnvfOc7WbNmjUP+SlIJZsyYwf79+2k0\nGjzyyCM8/vjjtW7fcJekLnrHO97Bddddx+7du2vdruEuSR12/PhxXnjhBQBefvll9uzZwyWXXFLr\nPnxaRlLfG+/Rxbo9++yz3HzzzQwPD/Paa69x44038sEPfrDWfRjuktRhl19+Ofv27WvrPuyWkaQC\nGe6SVCDDXVJfavdfQpqqqdZnuEvqO3PmzOHEiRPTNuAzkxMnTjBnzpxJb8MbqpL6zuDgII1Gg+n8\n5z7nzJnD4ODgpL/fcJfUd2bNmsXixYu7XUZb2S0jSQUy3CWpQIa7JBXIcJekAhnuklQgw12SCmS4\nS1KBDHdJKpDhLkkFMtwlqUCGuyQVyHCXpAIZ7pJUIMNdkgpkuEtSgQx3SSqQ4S5JBTLcJalAhrsk\nFchwl6QCGe6SVKBK4R4RayPiYEQciojt51j/OxHxWETsj4j/iYil9ZcqSapq3HCPiBnAvcA6YClw\n0znC+yuZeVlmLgc+C3yu9kolSZVVuXK/CjiUmU9n5qvADmBja4PM/EnL7FuBrK9ESdJEzazQZgFw\npGW+AVw9ulFE3AbcDswGfqWW6iRJk1LbDdXMvDcz3wv8AfDH52oTEVsiYigiho4fP17XriVJo1QJ\n96PAwpb5weaysewAPnSuFZl5X2auzMyVAwMD1auUJE1IlXDfCyyJiMURMRvYDOxsbRARS1pmfw14\nqr4SJUkTNW6fe2aejoitwEPADOCLmXkgIu4EhjJzJ7A1It4PnAJ+BNzczqIlSW+uyg1VMnMXsGvU\nsjtaprfVXJckaQp8Q1WSCmS4S1KBDHdJKpDhLkkFMtwlqUCGuyQVyHCXpAIZ7pJUIMNdkgpkuEtS\ngQx3SSqQ4S5JBTLcJalAhrskFchwl6QCGe6SVCDDXZIKZLhLUoEMd0kqkOEuSQUy3CWpQIa7JBXI\ncJekAhnuklQgw12SCmS4S1KBDHdJKpDhLkkFMtwlqUCGuyQVyHCXpAIZ7pJUIMNdkgpkuEtSgQx3\nSSqQ4S5JBTLcJalAlcI9ItZGxMGIOBQR28+x/vaIeCIivhMRX4uI99RfqiSpqnHDPSJmAPcC64Cl\nwE0RsXRUs33Aysy8HPgq8Nm6C5UkVVflyv0q4FBmPp2ZrwI7gI2tDTLz65n50+bsN4HBesuUJE1E\nlXBfABxpmW80l43lFuDfp1KUJGlqZta5sYj4DWAl8MtjrN8CbAG48MIL69y1JKlFlSv3o8DClvnB\n5rKzRMT7gT8CNmTmK+faUGbel5krM3PlwMDAZOqVJFVQJdz3AksiYnFEzAY2AztbG0TEFcDnGQn2\n5+ovU5I0EeOGe2aeBrYCDwFPAg9k5oGIuDMiNjSb/QUwF/jHiNgfETvH2JwkqQMq9bln5i5g16hl\nd7RMv7/muiRJU+AbqpJUIMNdkgpkuEtSgQx3SSqQ4S5JBTLcJalAhrskFchwl6QCGe6SVCDDXZIK\nZLhLUoEMd0kqkOEuSQUy3CWpQIa7JBXIcJekAhnuklQgw12SCmS4S1KBDHdJKpDhLkkFMtwlqUCG\nuyQVyHCXpAIZ7pJUIMNdkgpkuEtSgQx3SSqQ4S5JBTLcJalAhrskFchwl6QCGe6SVCDDXZIKZLhL\nUoEMd0kqkOEuSQWqFO4RsTYiDkbEoYjYfo7110bEtyPidETcUH+ZkqSJmDleg4iYAdwLrAEawN6I\n2JmZT7Q0+z7wW8DvtaNISZ134MSBSu2WzV/W5ko0GeOGO3AVcCgznwaIiB3ARuD1cM/Mw811r7Wh\nRknSBFXpllkAHGmZbzSXSZKmqY7eUI2ILRExFBFDx48f7+SuJamvVAn3o8DClvnB5rIJy8z7MnNl\nZq4cGBiYzCYkSRVUCfe9wJKIWBwRs4HNwM72liVJmopxwz0zTwNbgYeAJ4EHMvNARNwZERsAImJV\nRDSADwOfj4hqt9klSW1R5WkZMnMXsGvUsjtapvcy0l0jSZoGfENVkgpkuEtSgSp1y0jSeKq+0Qq+\n1doJXrlLUoEMd0kqkOEuSQWyz11jclRAqXcZ7pLO7fjBau385T4t2S0jSQXyyl1d5eNzHfKDfdXa\nvfuK9tahjjHcJdXi0LGXKrddNr+NhQgw3KXe06Gr8CM/fLlSOz9PTU/2uUtSgbxyV63mPP+dag0H\nf6m9hajjznvh0ARaXz35HU3mk0sf3nMw3KU6TTREqrZv/Z5JOPCTZyq1W9YD4TaZYynp+Ksy3FWr\n7/70+5XaLWlzHeoNPi3VPoa7pO6p+qIU+LLUBBnu0lj6sJ9W5TDc1R86FNSd6Nvtx/5jTZzhru7y\nY7nUFoa7uqrqizLQ8rJMQd0lVa/CwStxTYzhLo3B7o8+1+MXEYa7xlR1rJBeGCfEoC7HZIZF6Meh\nFAx39RyDuhyT6pZTJYa7xlT9dfIpvEouTVO9fhHhwGGSVCDDXZIKZLhLUoEMd0kqkDdUpRpN10fu\npmtdRenQ8M1V9Wa4T7N/RJXJQOxvvf7/35vhLk3QdP1Bncxz3tP1WDS9GO7qOYabpqPpNk5QT4b7\npP4Re3ycCEmaCJ+WkaQC9eSVuybBTy5SX+mbcJ/UOBEGoqSKptsgaJXCPSLWAn8FzAD+JjP/fNT6\n84C/A64ETgAfyczD9Zb6M9PtH7EX9PogSJImZtxwj4gZwL3AGqAB7I2InZn5REuzW4AfZebFEbEZ\n+AzwkXYU3Ele7UvqVVWu3K8CDmXm0wARsQPYCLSG+0bgk83prwL3RERkZtZY65R06vG5if5COPC/\n/1x528suuX5kYhK/QHx8UOovVcJ9AXCkZb7BGwfwfr1NZp6OiB8D84Hn6yiyW6brX3zZffiJ8RsB\na/10IPWtGO/iOiJuANZm5q3N+d8Ers7MrS1tHm+2aTTnv9ts8/yobW0BtjRnfwE4WNeBAOfT479M\npsjj9/g9/v7wnswcGK9RlSv3o8DClvnB5rJztWlExEzg7YzcWD1LZt4H3FdhnxMWEUOZubId2+4F\nHr/H7/H37/GfS5WXmPYCSyJicUTMBjYDO0e12Qnc3Jy+AXh4OvW3S1K/GffKvdmHvhV4iJFHIb+Y\nmQci4k5gKDN3An8L3B8Rh4AfMvILQJLUJZWec8/MXcCuUcvuaJk+CXy43tImrC3dPT3E4+9vHr/O\nMu4NVUlS73HgMEkqUBHhHhFrI+JgRByKiO3drqfTIuJwRDwWEfsjYqjb9bRbRHwxIp5rPoJ7ZtnP\nR8SeiHiq+fXnulljO41x/J+MiKPNc2B/RKzvZo3tFBELI+LrEfFERByIiG3N5X1zDlTR8+HeMjzC\nOmApcFNELO1uVV1xXWYu75PHwb4ErB21bDvwtcxcAnytOV+qL/HG4wf4y+Y5sLx5n6xUp4FPZOZS\n4BrgtubPfD+dA+Pq+XCnZXiEzHwVODM8ggqVmf/FyFNZrTYCX25Ofxn4UEeL6qAxjr9vZOazmfnt\n5vSLwJOMvCXfN+dAFSWE+7mGR1jQpVq6JYH/iIhHm28B96N3Zeazzen/A97VzWK6ZGtEfKfZbdMX\nXRIRsQi4AvgWngNnKSHcBaszcwUjXVO3RcS13S6om5ov0PXbY2B/DbwXWA48C9zV3XLaLyLmAv8E\n/G5m/qR1XZ+eA2cpIdyrDI9QtMw82vz6HPAvjHRV9ZtjEXEBQPPrc12up6My81hmDmfma8AXKPwc\niIhZjAT732fmmaFV+/ocGK2EcK8yPEKxIuKtETHvzDTwq8Djb/5dRWodAuNm4MEu1tJxZ0Kt6dcp\n+ByIiGDkrfgnM/NzLav6+hwYrYiXmJqPfd3Nz4ZH+LMul9QxEXERI1frMPLG8VdKP/6I+AfgfYyM\nBHgM+BPgX4EHgAuB7wE3ZmaRNx3HOP73MdIlk8Bh4Ldb+p+LEhGrgf8GHgNeay7+Q0b63fviHKii\niHCXJJ2thG4ZSdIohrskFchwl6QCGe6SVCDDXZIKZLhLUoEMd0kqkOEuSQX6f7R/gdnOlUs4AAAA\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x349261210>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(nclusters):\n",
    "    if i in [1]: continue\n",
    "        \n",
    "    distr = paxsample[pax_kmeans.labels_ == i].sum(axis=0) / paxsample[pax_kmeans.labels_ == i].sum(axis=0).sum()\n",
    "    plt.bar(range(24),distr, alpha=0.2, label=str(i))\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 368,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([  7.19074000e+05,   2.57838000e+05,   1.33386000e+05,\n",
       "          7.86160000e+04,   5.13490000e+04,   3.49960000e+04,\n",
       "          2.46020000e+04,   1.78010000e+04,   1.33580000e+04,\n",
       "          9.83500000e+03,   7.42800000e+03,   5.44700000e+03,\n",
       "          3.97300000e+03,   2.88200000e+03,   2.11300000e+03,\n",
       "          1.49600000e+03,   1.09400000e+03,   7.52000000e+02,\n",
       "          5.78000000e+02,   3.96000000e+02,   3.09000000e+02,\n",
       "          2.00000000e+02,   1.33000000e+02,   9.10000000e+01,\n",
       "          7.80000000e+01,   5.80000000e+01,   4.20000000e+01,\n",
       "          2.90000000e+01,   1.40000000e+01,   1.30000000e+01,\n",
       "          1.50000000e+01,   1.60000000e+01,   1.00000000e+01,\n",
       "          9.00000000e+00,   7.00000000e+00,   4.00000000e+00,\n",
       "          5.00000000e+00,   4.00000000e+00,   2.00000000e+00,\n",
       "          1.00000000e+00,   1.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   1.00000000e+00,   2.00000000e+00,\n",
       "          1.00000000e+00,   1.00000000e+00,   1.00000000e+00,\n",
       "          1.00000000e+00,   1.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          1.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   1.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   1.00000000e+00,\n",
       "          0.00000000e+00,   1.00000000e+00,   0.00000000e+00,\n",
       "          1.00000000e+00,   1.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          0.00000000e+00,   0.00000000e+00,   0.00000000e+00,\n",
       "          1.00000000e+00]),\n",
       " array([   1.,    4.,    7.,   10.,   13.,   16.,   19.,   22.,   25.,\n",
       "          28.,   31.,   34.,   37.,   40.,   43.,   46.,   49.,   52.,\n",
       "          55.,   58.,   61.,   64.,   67.,   70.,   73.,   76.,   79.,\n",
       "          82.,   85.,   88.,   91.,   94.,   97.,  100.,  103.,  106.,\n",
       "         109.,  112.,  115.,  118.,  121.,  124.,  127.,  130.,  133.,\n",
       "         136.,  139.,  142.,  145.,  148.,  151.,  154.,  157.,  160.,\n",
       "         163.,  166.,  169.,  172.,  175.,  178.,  181.,  184.,  187.,\n",
       "         190.,  193.,  196.,  199.,  202.,  205.,  208.,  211.,  214.,\n",
       "         217.,  220.,  223.,  226.,  229.,  232.,  235.,  238.,  241.,\n",
       "         244.,  247.,  250.,  253.,  256.,  259.,  262.,  265.,  268.,\n",
       "         271.,  274.,  277.,  280.,  283.,  286.,  289.,  292.,  295.,\n",
       "         298.,  301.]),\n",
       " <a list of 100 Patch objects>)"
      ]
     },
     "execution_count": 368,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAD8CAYAAACLrvgBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x36fa23990>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(paxsums[paxsums != paxsums.max()], bins = 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}