CannedGeo_/Facies_classification-BPage_CannedGeo_F1_56-VALIDATED.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Facies classification using Machine Learning\n",
    "\n",
    "<hr />\n",
    "\n",
    "## Contest entry by Bryan Page\n",
    "\n",
    "#### This version has been validated by Matt\n",
    "\n",
    "<hr />\n",
    "\n",
    "### Based on [the original notebook](../Facies_classification.ipynb) by Brendon Hall, [Enthought](https://www.enthought.com/)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr />\n",
    "\n",
    "### Matt's current sklearn version"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.18\n"
     ]
    }
   ],
   "source": [
    "import sklearn\n",
    "\n",
    "print(sklearn.__version__)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr />\n",
    "\n",
    "## Exploring the dataset\n",
    "\n",
    "First, we will examine the data set we will use to train the classifier."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>Facies</th>\n",
       "      <th>Formation</th>\n",
       "      <th>Well Name</th>\n",
       "      <th>Depth</th>\n",
       "      <th>GR</th>\n",
       "      <th>ILD_log10</th>\n",
       "      <th>DeltaPHI</th>\n",
       "      <th>PHIND</th>\n",
       "      <th>PE</th>\n",
       "      <th>NM_M</th>\n",
       "      <th>RELPOS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>A1 SH</td>\n",
       "      <td>SHRIMPLIN</td>\n",
       "      <td>2793.0</td>\n",
       "      <td>77.450</td>\n",
       "      <td>0.664</td>\n",
       "      <td>9.900</td>\n",
       "      <td>11.915</td>\n",
       "      <td>4.600</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>A1 SH</td>\n",
       "      <td>SHRIMPLIN</td>\n",
       "      <td>2793.5</td>\n",
       "      <td>78.260</td>\n",
       "      <td>0.661</td>\n",
       "      <td>14.200</td>\n",
       "      <td>12.565</td>\n",
       "      <td>4.100</td>\n",
       "      <td>1</td>\n",
       "      <td>0.979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>A1 SH</td>\n",
       "      <td>SHRIMPLIN</td>\n",
       "      <td>2794.0</td>\n",
       "      <td>79.050</td>\n",
       "      <td>0.658</td>\n",
       "      <td>14.800</td>\n",
       "      <td>13.050</td>\n",
       "      <td>3.600</td>\n",
       "      <td>1</td>\n",
       "      <td>0.957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>A1 SH</td>\n",
       "      <td>SHRIMPLIN</td>\n",
       "      <td>2794.5</td>\n",
       "      <td>86.100</td>\n",
       "      <td>0.655</td>\n",
       "      <td>13.900</td>\n",
       "      <td>13.115</td>\n",
       "      <td>3.500</td>\n",
       "      <td>1</td>\n",
       "      <td>0.936</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>A1 SH</td>\n",
       "      <td>SHRIMPLIN</td>\n",
       "      <td>2795.0</td>\n",
       "      <td>74.580</td>\n",
       "      <td>0.647</td>\n",
       "      <td>13.500</td>\n",
       "      <td>13.300</td>\n",
       "      <td>3.400</td>\n",
       "      <td>1</td>\n",
       "      <td>0.915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4144</th>\n",
       "      <td>5</td>\n",
       "      <td>C LM</td>\n",
       "      <td>CHURCHMAN BIBLE</td>\n",
       "      <td>3120.5</td>\n",
       "      <td>46.719</td>\n",
       "      <td>0.947</td>\n",
       "      <td>1.828</td>\n",
       "      <td>7.254</td>\n",
       "      <td>3.617</td>\n",
       "      <td>2</td>\n",
       "      <td>0.685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4145</th>\n",
       "      <td>5</td>\n",
       "      <td>C LM</td>\n",
       "      <td>CHURCHMAN BIBLE</td>\n",
       "      <td>3121.0</td>\n",
       "      <td>44.563</td>\n",
       "      <td>0.953</td>\n",
       "      <td>2.241</td>\n",
       "      <td>8.013</td>\n",
       "      <td>3.344</td>\n",
       "      <td>2</td>\n",
       "      <td>0.677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4146</th>\n",
       "      <td>5</td>\n",
       "      <td>C LM</td>\n",
       "      <td>CHURCHMAN BIBLE</td>\n",
       "      <td>3121.5</td>\n",
       "      <td>49.719</td>\n",
       "      <td>0.964</td>\n",
       "      <td>2.925</td>\n",
       "      <td>8.013</td>\n",
       "      <td>3.190</td>\n",
       "      <td>2</td>\n",
       "      <td>0.669</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4147</th>\n",
       "      <td>5</td>\n",
       "      <td>C LM</td>\n",
       "      <td>CHURCHMAN BIBLE</td>\n",
       "      <td>3122.0</td>\n",
       "      <td>51.469</td>\n",
       "      <td>0.965</td>\n",
       "      <td>3.083</td>\n",
       "      <td>7.708</td>\n",
       "      <td>3.152</td>\n",
       "      <td>2</td>\n",
       "      <td>0.661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4148</th>\n",
       "      <td>5</td>\n",
       "      <td>C LM</td>\n",
       "      <td>CHURCHMAN BIBLE</td>\n",
       "      <td>3122.5</td>\n",
       "      <td>50.031</td>\n",
       "      <td>0.970</td>\n",
       "      <td>2.609</td>\n",
       "      <td>6.668</td>\n",
       "      <td>3.295</td>\n",
       "      <td>2</td>\n",
       "      <td>0.653</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4149 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Facies Formation        Well Name   Depth      GR  ILD_log10  DeltaPHI  \\\n",
       "0          3     A1 SH        SHRIMPLIN  2793.0  77.450      0.664     9.900   \n",
       "1          3     A1 SH        SHRIMPLIN  2793.5  78.260      0.661    14.200   \n",
       "2          3     A1 SH        SHRIMPLIN  2794.0  79.050      0.658    14.800   \n",
       "3          3     A1 SH        SHRIMPLIN  2794.5  86.100      0.655    13.900   \n",
       "4          3     A1 SH        SHRIMPLIN  2795.0  74.580      0.647    13.500   \n",
       "...      ...       ...              ...     ...     ...        ...       ...   \n",
       "4144       5      C LM  CHURCHMAN BIBLE  3120.5  46.719      0.947     1.828   \n",
       "4145       5      C LM  CHURCHMAN BIBLE  3121.0  44.563      0.953     2.241   \n",
       "4146       5      C LM  CHURCHMAN BIBLE  3121.5  49.719      0.964     2.925   \n",
       "4147       5      C LM  CHURCHMAN BIBLE  3122.0  51.469      0.965     3.083   \n",
       "4148       5      C LM  CHURCHMAN BIBLE  3122.5  50.031      0.970     2.609   \n",
       "\n",
       "       PHIND     PE  NM_M  RELPOS  \n",
       "0     11.915  4.600     1   1.000  \n",
       "1     12.565  4.100     1   0.979  \n",
       "2     13.050  3.600     1   0.957  \n",
       "3     13.115  3.500     1   0.936  \n",
       "4     13.300  3.400     1   0.915  \n",
       "...      ...    ...   ...     ...  \n",
       "4144   7.254  3.617     2   0.685  \n",
       "4145   8.013  3.344     2   0.677  \n",
       "4146   8.013  3.190     2   0.669  \n",
       "4147   7.708  3.152     2   0.661  \n",
       "4148   6.668  3.295     2   0.653  \n",
       "\n",
       "[4149 rows x 11 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.colors as colors\n",
    "from mpl_toolkits.axes_grid1 import make_axes_locatable\n",
    "\n",
    "from pandas import set_option\n",
    "set_option(\"display.max_rows\", 10)\n",
    "pd.options.mode.chained_assignment = None\n",
    "\n",
    "filename = '../facies_vectors.csv'\n",
    "training_data = pd.read_csv(filename)\n",
    "training_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Remove a single well to use as a blind test later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "blind = training_data[training_data['Well Name'] == 'SHANKLE']\n",
    "training_data = training_data[training_data['Well Name'] != 'SHANKLE']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'SHANKLE' in training_data['Well Name'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[SHRIMPLIN, ALEXANDER D, LUKE G U, KIMZEY A, CROSS H CATTLE, NOLAN, Recruit F9, NEWBY, CHURCHMAN BIBLE]\n",
       "Categories (9, object): [SHRIMPLIN, ALEXANDER D, LUKE G U, KIMZEY A, ..., NOLAN, Recruit F9, NEWBY, CHURCHMAN BIBLE]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data['Well Name'] = training_data['Well Name'].astype('category')\n",
    "training_data['Formation'] = training_data['Formation'].astype('category')\n",
    "training_data['Well Name'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These are the names of the 10 training wells in the Council Grove reservoir.  Data has been recruited into pseudo-well 'Recruit F9' to better represent facies 9, the Phylloid-algal bafflestone. \n",
    "\n",
    "Before we plot the well data, let's define a color map so the facies are represented by consistent color in all the plots in this tutorial.  We also create the abbreviated facies labels, and add those to the `facies_vectors` dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/matt/anaconda/envs/python3/lib/python3.5/site-packages/numpy/lib/function_base.py:3834: RuntimeWarning: Invalid value encountered in percentile\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Facies</th>\n",
       "      <th>Depth</th>\n",
       "      <th>GR</th>\n",
       "      <th>ILD_log10</th>\n",
       "      <th>DeltaPHI</th>\n",
       "      <th>PHIND</th>\n",
       "      <th>PE</th>\n",
       "      <th>NM_M</th>\n",
       "      <th>RELPOS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3700.000000</td>\n",
       "      <td>3700.000000</td>\n",
       "      <td>3700.000000</td>\n",
       "      <td>3700.000000</td>\n",
       "      <td>3700.000000</td>\n",
       "      <td>3700.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>3700.000000</td>\n",
       "      <td>3700.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.615676</td>\n",
       "      <td>2908.853378</td>\n",
       "      <td>64.873649</td>\n",
       "      <td>0.663053</td>\n",
       "      <td>4.651677</td>\n",
       "      <td>12.892826</td>\n",
       "      <td>3.805693</td>\n",
       "      <td>1.540000</td>\n",
       "      <td>0.524125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.475808</td>\n",
       "      <td>139.010855</td>\n",
       "      <td>30.817166</td>\n",
       "      <td>0.253863</td>\n",
       "      <td>5.109006</td>\n",
       "      <td>6.796219</td>\n",
       "      <td>0.894118</td>\n",
       "      <td>0.498465</td>\n",
       "      <td>0.287147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>2573.500000</td>\n",
       "      <td>10.149000</td>\n",
       "      <td>-0.025949</td>\n",
       "      <td>-21.832000</td>\n",
       "      <td>0.550000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>2818.500000</td>\n",
       "      <td>43.778250</td>\n",
       "      <td>0.502000</td>\n",
       "      <td>1.800000</td>\n",
       "      <td>8.350000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>2939.000000</td>\n",
       "      <td>64.817000</td>\n",
       "      <td>0.645613</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>11.857500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.531000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>7.000000</td>\n",
       "      <td>3015.125000</td>\n",
       "      <td>80.322500</td>\n",
       "      <td>0.823000</td>\n",
       "      <td>7.600000</td>\n",
       "      <td>15.750000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.772000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>9.000000</td>\n",
       "      <td>3138.000000</td>\n",
       "      <td>361.150000</td>\n",
       "      <td>1.800000</td>\n",
       "      <td>19.312000</td>\n",
       "      <td>84.400000</td>\n",
       "      <td>8.094000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Facies        Depth           GR    ILD_log10     DeltaPHI  \\\n",
       "count  3700.000000  3700.000000  3700.000000  3700.000000  3700.000000   \n",
       "mean      4.615676  2908.853378    64.873649     0.663053     4.651677   \n",
       "std       2.475808   139.010855    30.817166     0.253863     5.109006   \n",
       "min       1.000000  2573.500000    10.149000    -0.025949   -21.832000   \n",
       "25%       2.000000  2818.500000    43.778250     0.502000     1.800000   \n",
       "50%       4.000000  2939.000000    64.817000     0.645613     4.400000   \n",
       "75%       7.000000  3015.125000    80.322500     0.823000     7.600000   \n",
       "max       9.000000  3138.000000   361.150000     1.800000    19.312000   \n",
       "\n",
       "             PHIND           PE         NM_M       RELPOS  \n",
       "count  3700.000000  2783.000000  3700.000000  3700.000000  \n",
       "mean     12.892826     3.805693     1.540000     0.524125  \n",
       "std       6.796219     0.894118     0.498465     0.287147  \n",
       "min       0.550000     0.200000     1.000000     0.000000  \n",
       "25%       8.350000          NaN     1.000000     0.278000  \n",
       "50%      11.857500          NaN     2.000000     0.531000  \n",
       "75%      15.750000          NaN     2.000000     0.772000  \n",
       "max      84.400000     8.094000     2.000000     1.000000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1=sandstone  2=c_siltstone   3=f_siltstone # 4=marine_silt_shale \n",
    "#5=mudstone 6=wackestone 7=dolomite 8=packstone 9=bafflestone\n",
    "facies_colors = ['#F4D03F', '#F5B041', '#DC7633','#A569BD',\n",
    "       '#000000', '#000080', '#2E86C1', '#AED6F1', '#196F3D']\n",
    "\n",
    "facies_labels = ['SS', 'CSiS', 'FSiS', 'SiSh', 'MS',\n",
    "                 'WS', 'D','PS', 'BS']\n",
    "#facies_color_map is a dictionary that maps facies labels\n",
    "#to their respective colors\n",
    "facies_color_map = {}\n",
    "for ind, label in enumerate(facies_labels):\n",
    "    facies_color_map[label] = facies_colors[ind]\n",
    "\n",
    "def label_facies(row, labels):\n",
    "    return labels[ row['Facies'] -1]\n",
    "    \n",
    "training_data.loc[:,'FaciesLabels'] = training_data.apply(lambda row: label_facies(row, facies_labels), axis=1)\n",
    "training_data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is a quick view of the statistical distribution of the input variables.  Looking at the `count` values, most values have 4149 valid values except for `PE`, which has 3232.  In this tutorial we will drop the feature vectors that don't have a valid `PE` entry."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "PE_mask = training_data['PE'].notnull().values\n",
    "training_data = training_data[PE_mask]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>Facies</th>\n",
       "      <th>Depth</th>\n",
       "      <th>GR</th>\n",
       "      <th>ILD_log10</th>\n",
       "      <th>DeltaPHI</th>\n",
       "      <th>PHIND</th>\n",
       "      <th>PE</th>\n",
       "      <th>NM_M</th>\n",
       "      <th>RELPOS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.558390</td>\n",
       "      <td>2873.456522</td>\n",
       "      <td>66.249445</td>\n",
       "      <td>0.644637</td>\n",
       "      <td>3.754963</td>\n",
       "      <td>13.118929</td>\n",
       "      <td>3.805693</td>\n",
       "      <td>1.523895</td>\n",
       "      <td>0.523057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.515249</td>\n",
       "      <td>138.349383</td>\n",
       "      <td>31.610849</td>\n",
       "      <td>0.241922</td>\n",
       "      <td>5.045916</td>\n",
       "      <td>7.389665</td>\n",
       "      <td>0.894118</td>\n",
       "      <td>0.499518</td>\n",
       "      <td>0.287499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>2573.500000</td>\n",
       "      <td>13.250000</td>\n",
       "      <td>-0.025949</td>\n",
       "      <td>-21.832000</td>\n",
       "      <td>0.550000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.010000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>2767.500000</td>\n",
       "      <td>46.081500</td>\n",
       "      <td>0.497000</td>\n",
       "      <td>1.300000</td>\n",
       "      <td>8.165000</td>\n",
       "      <td>3.200000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.276500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>2895.000000</td>\n",
       "      <td>65.541000</td>\n",
       "      <td>0.627000</td>\n",
       "      <td>3.581000</td>\n",
       "      <td>11.900000</td>\n",
       "      <td>3.600000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.529000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>7.000000</td>\n",
       "      <td>2987.000000</td>\n",
       "      <td>80.714000</td>\n",
       "      <td>0.812323</td>\n",
       "      <td>6.500000</td>\n",
       "      <td>16.144000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.771500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>9.000000</td>\n",
       "      <td>3122.500000</td>\n",
       "      <td>361.150000</td>\n",
       "      <td>1.480000</td>\n",
       "      <td>18.500000</td>\n",
       "      <td>84.400000</td>\n",
       "      <td>8.094000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Facies        Depth           GR    ILD_log10     DeltaPHI  \\\n",
       "count  2783.000000  2783.000000  2783.000000  2783.000000  2783.000000   \n",
       "mean      4.558390  2873.456522    66.249445     0.644637     3.754963   \n",
       "std       2.515249   138.349383    31.610849     0.241922     5.045916   \n",
       "min       1.000000  2573.500000    13.250000    -0.025949   -21.832000   \n",
       "25%       2.000000  2767.500000    46.081500     0.497000     1.300000   \n",
       "50%       4.000000  2895.000000    65.541000     0.627000     3.581000   \n",
       "75%       7.000000  2987.000000    80.714000     0.812323     6.500000   \n",
       "max       9.000000  3122.500000   361.150000     1.480000    18.500000   \n",
       "\n",
       "             PHIND           PE         NM_M       RELPOS  \n",
       "count  2783.000000  2783.000000  2783.000000  2783.000000  \n",
       "mean     13.118929     3.805693     1.523895     0.523057  \n",
       "std       7.389665     0.894118     0.499518     0.287499  \n",
       "min       0.550000     0.200000     1.000000     0.010000  \n",
       "25%       8.165000     3.200000     1.000000     0.276500  \n",
       "50%      11.900000     3.600000     2.000000     0.529000  \n",
       "75%      16.144000     4.400000     2.000000     0.771500  \n",
       "max      84.400000     8.094000     2.000000     1.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's take a look at the data from individual wells in a more familiar log plot form.  We will create plots for the five well log variables, as well as a log for facies labels.  The plots are based on the those described in Alessandro Amato del Monte's [excellent tutorial](https://github.com/seg/tutorials/tree/master/1504_Seismic_petrophysics_1)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#I changed the colors to match the paper for easier compare.\n",
    "def make_facies_log_plot(logs, facies_colors):\n",
    "    #make sure logs are sorted by depth\n",
    "    logs = logs.sort_values(by='Depth')\n",
    "    cmap_facies = colors.ListedColormap(\n",
    "            facies_colors[0:len(facies_colors)], 'indexed')\n",
    "    \n",
    "    ztop=logs.Depth.min(); zbot=logs.Depth.max()\n",
    "    \n",
    "    cluster=np.repeat(np.expand_dims(logs['Facies'].values,1), 100, 1)\n",
    "    \n",
    "    f, ax = plt.subplots(nrows=1, ncols=6, figsize=(8, 12))\n",
    "    ax[0].plot(logs.GR, logs.Depth, '-g')\n",
    "    ax[1].plot(logs.ILD_log10, logs.Depth, '-')\n",
    "    ax[2].plot(logs.DeltaPHI, logs.Depth, '-', color='0.5')\n",
    "    ax[3].plot(logs.PHIND, logs.Depth, '-', color='r')\n",
    "    ax[4].plot(logs.PE, logs.Depth, '-', color='black')\n",
    "    im=ax[5].imshow(cluster, interpolation='none', aspect='auto',\n",
    "                    cmap=cmap_facies,vmin=1,vmax=9)\n",
    "    \n",
    "    divider = make_axes_locatable(ax[5])\n",
    "    cax = divider.append_axes(\"right\", size=\"20%\", pad=0.05)\n",
    "    cbar=plt.colorbar(im, cax=cax)\n",
    "    cbar.set_label((17*' ').join([' SS ', 'CSiS', 'FSiS', \n",
    "                                'SiSh', ' MS ', ' WS ', ' D  ', \n",
    "                                ' PS ', ' BS ']))\n",
    "    cbar.set_ticks(range(0,1)); cbar.set_ticklabels('')\n",
    "    \n",
    "    for i in range(len(ax)-1):\n",
    "        ax[i].set_ylim(ztop,zbot)\n",
    "        ax[i].invert_yaxis()\n",
    "        ax[i].grid()\n",
    "        ax[i].locator_params(axis='x', nbins=3)\n",
    "    \n",
    "    ax[0].set_xlabel(\"GR\")\n",
    "    ax[0].set_xlim(logs.GR.min(),logs.GR.max())\n",
    "    ax[1].set_xlabel(\"ILD_log10\")\n",
    "    ax[1].set_xlim(logs.ILD_log10.min(),logs.ILD_log10.max())\n",
    "    ax[2].set_xlabel(\"DeltaPHI\")\n",
    "    ax[2].set_xlim(logs.DeltaPHI.min(),logs.DeltaPHI.max())\n",
    "    ax[3].set_xlabel(\"PHIND\")\n",
    "    ax[3].set_xlim(logs.PHIND.min(),logs.PHIND.max())\n",
    "    ax[4].set_xlabel(\"PE\")\n",
    "    ax[4].set_xlim(logs.PE.min(),logs.PE.max())\n",
    "    ax[5].set_xlabel('Facies')\n",
    "    \n",
    "    ax[1].set_yticklabels([]); ax[2].set_yticklabels([]); ax[3].set_yticklabels([])\n",
    "    ax[4].set_yticklabels([]); ax[5].set_yticklabels([])\n",
    "    ax[5].set_xticklabels([])\n",
    "    f.suptitle('Well: %s'%logs.iloc[0]['Well Name'], fontsize=14,y=0.94)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In addition to individual wells, we can look at how the various facies are represented by the entire training set.  Let's plot a histgram of the number of training examples for each facies class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SS      170\n",
       "CSiS    649\n",
       "FSiS    498\n",
       "SiSh    177\n",
       "MS      198\n",
       "WS      391\n",
       "D        81\n",
       "PS      458\n",
       "BS      161\n",
       "Name: Facies, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEWCAYAAAB2X2wCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAYAs3lIIkwI1AAAAAElEQVQ1rwYeAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10f3537b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#count the number of unique entries for each facies, sort them by\n",
    "#facies number (instead of by number of entries)\n",
    "facies_counts = training_data['Facies'].value_counts().sort_index()\n",
    "#use facies labels to index each count\n",
    "facies_counts.index = facies_labels\n",
    "\n",
    "facies_counts.plot(kind='bar',color=facies_colors, \n",
    "                   title='Distribution of Training Data by Facies')\n",
    "facies_counts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conditioning the data set\n",
    "\n",
    "Now we extract just the feature variables we need to perform the classification.  The predictor variables are the five wireline values and two geologic constraining variables. We also get a vector of the facies labels that correspond to each feature vector."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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></th>\n",
       "      <th>GR</th>\n",
       "      <th>ILD_log10</th>\n",
       "      <th>DeltaPHI</th>\n",
       "      <th>PHIND</th>\n",
       "      <th>PE</th>\n",
       "      <th>NM_M</th>\n",
       "      <th>RELPOS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>2783.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>66.249445</td>\n",
       "      <td>0.644637</td>\n",
       "      <td>3.754963</td>\n",
       "      <td>13.118929</td>\n",
       "      <td>3.805693</td>\n",
       "      <td>1.523895</td>\n",
       "      <td>0.523057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>31.610849</td>\n",
       "      <td>0.241922</td>\n",
       "      <td>5.045916</td>\n",
       "      <td>7.389665</td>\n",
       "      <td>0.894118</td>\n",
       "      <td>0.499518</td>\n",
       "      <td>0.287499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>13.250000</td>\n",
       "      <td>-0.025949</td>\n",
       "      <td>-21.832000</td>\n",
       "      <td>0.550000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.010000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>46.081500</td>\n",
       "      <td>0.497000</td>\n",
       "      <td>1.300000</td>\n",
       "      <td>8.165000</td>\n",
       "      <td>3.200000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.276500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>65.541000</td>\n",
       "      <td>0.627000</td>\n",
       "      <td>3.581000</td>\n",
       "      <td>11.900000</td>\n",
       "      <td>3.600000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.529000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>80.714000</td>\n",
       "      <td>0.812323</td>\n",
       "      <td>6.500000</td>\n",
       "      <td>16.144000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.771500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>361.150000</td>\n",
       "      <td>1.480000</td>\n",
       "      <td>18.500000</td>\n",
       "      <td>84.400000</td>\n",
       "      <td>8.094000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                GR    ILD_log10     DeltaPHI        PHIND           PE  \\\n",
       "count  2783.000000  2783.000000  2783.000000  2783.000000  2783.000000   \n",
       "mean     66.249445     0.644637     3.754963    13.118929     3.805693   \n",
       "std      31.610849     0.241922     5.045916     7.389665     0.894118   \n",
       "min      13.250000    -0.025949   -21.832000     0.550000     0.200000   \n",
       "25%      46.081500     0.497000     1.300000     8.165000     3.200000   \n",
       "50%      65.541000     0.627000     3.581000    11.900000     3.600000   \n",
       "75%      80.714000     0.812323     6.500000    16.144000     4.400000   \n",
       "max     361.150000     1.480000    18.500000    84.400000     8.094000   \n",
       "\n",
       "              NM_M       RELPOS  \n",
       "count  2783.000000  2783.000000  \n",
       "mean      1.523895     0.523057  \n",
       "std       0.499518     0.287499  \n",
       "min       1.000000     0.010000  \n",
       "25%       1.000000     0.276500  \n",
       "50%       2.000000     0.529000  \n",
       "75%       2.000000     0.771500  \n",
       "max       2.000000     1.000000  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "correct_facies_labels = training_data['Facies'].values\n",
    "\n",
    "feature_vectors = training_data.drop(['Formation', 'Well Name', 'Depth','Facies','FaciesLabels'], axis=1)\n",
    "feature_vectors.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import preprocessing\n",
    "\n",
    "scaler = preprocessing.StandardScaler().fit(feature_vectors)\n",
    "scaled_features = scaler.transform(feature_vectors)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/matt/anaconda/envs/python3/lib/python3.5/site-packages/sklearn/cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "        scaled_features, correct_facies_labels, test_size=0.3, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training the SVM classifier\n",
    "\n",
    "Now we use the cleaned and conditioned training set to create a facies classifier.  As mentioned above, we will use a type of machine learning model known as a [support vector machine](https://en.wikipedia.org/wiki/Support_vector_machine).  The SVM is a map of the feature vectors as points in a multi dimensional space, mapped so that examples from different facies are divided by a clear gap that is as wide as possible."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### [Matt: I changed this to the LinearSVC as well; I am also setting a random state to ensure it shuffles the same every time.]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import svm\n",
    "\n",
    "clf = svm.LinearSVC(random_state=42) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can train the classifier using the training set we created above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearSVC(C=1.0, class_weight=None, dual=True, fit_intercept=True,\n",
       "     intercept_scaling=1, loss='squared_hinge', max_iter=1000,\n",
       "     multi_class='ovr', penalty='l2', random_state=42, tol=0.0001,\n",
       "     verbose=0)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "predicted_labels = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Pred    SS  CSiS  FSiS  SiSh    MS    WS     D    PS    BS Total\n",
      "     True\n",
      "       SS    29    25     1                                        55\n",
      "     CSiS     3   160    33                             1         197\n",
      "     FSiS          48    79     1                       1     2   131\n",
      "     SiSh                 1    32          16           4     1    54\n",
      "       MS           3     1     8          45     2    14     2    75\n",
      "       WS                      12          65     1    29     2   109\n",
      "        D                       4                 3    19          26\n",
      "       PS           1     1     5          42          76    13   138\n",
      "       BS                       1           2          17    30    50\n",
      "\n",
      "Precision  0.91  0.68  0.68  0.51  0.00  0.38  0.50  0.47  0.60  0.54\n",
      "   Recall  0.53  0.81  0.60  0.59  0.00  0.60  0.12  0.55  0.60  0.57\n",
      "       F1  0.67  0.74  0.64  0.55  0.00  0.47  0.19  0.51  0.60  0.54\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "from classification_utilities import display_cm, display_adj_cm\n",
    "\n",
    "conf = confusion_matrix(y_test, predicted_labels)\n",
    "display_cm(conf, facies_labels, display_metrics=True, hide_zeros=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def accuracy(conf):\n",
    "    total_correct = 0.\n",
    "    nb_classes = conf.shape[0]\n",
    "    for i in np.arange(0,nb_classes):\n",
    "        total_correct += conf[i][i]\n",
    "    acc = total_correct/sum(sum(conf))\n",
    "    return acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "adjacent_facies = np.array([[1], [0,2], [1], [4], [3,5], [4,6,7], [5,7], [5,6,8], [6,7]])\n",
    "\n",
    "def accuracy_adjacent(conf, adjacent_facies):\n",
    "    nb_classes = conf.shape[0]\n",
    "    total_correct = 0.\n",
    "    for i in np.arange(0,nb_classes):\n",
    "        total_correct += conf[i][i]\n",
    "        for j in adjacent_facies[i]:\n",
    "            total_correct += conf[i][j]\n",
    "    return total_correct / sum(sum(conf))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Facies classification accuracy = 0.567665\n",
      "Adjacent facies classification accuracy = 0.906587\n"
     ]
    }
   ],
   "source": [
    "print('Facies classification accuracy = %f' % accuracy(conf))\n",
    "print('Adjacent facies classification accuracy = %f' % accuracy_adjacent(conf, adjacent_facies))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model parameter selection\n",
    "\n",
    "The classifier so far has been built with the default parameters.  However, we may be able to get improved classification results with optimal parameter choices."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#model selection takes a few minutes, change this variable\n",
    "#to true to run the parameter loop\n",
    "do_model_selection = False\n",
    "\n",
    "if do_model_selection:\n",
    "    C_range = np.array([.01, 1, 5, 10, 20, 50, 100, 1000, 5000, 10000])\n",
    "    gamma_range = np.array([0.0001, 0.001, 0.01, 0.1, 1, 10])\n",
    "    \n",
    "    fig, axes = plt.subplots(3, 2, \n",
    "                        sharex='col', sharey='row',figsize=(10,10))\n",
    "    plot_number = 0\n",
    "    for outer_ind, gamma_value in enumerate(gamma_range):\n",
    "        row = int(plot_number / 2)\n",
    "        column = int(plot_number % 2)\n",
    "        cv_errors = np.zeros(C_range.shape)\n",
    "        train_errors = np.zeros(C_range.shape)\n",
    "        for index, c_value in enumerate(C_range):\n",
    "            \n",
    "            clf = svm.SVC(C=c_value, gamma=gamma_value)\n",
    "            clf.fit(X_train,y_train)\n",
    "            \n",
    "            train_conf = confusion_matrix(y_train, clf.predict(X_train))\n",
    "            cv_conf = confusion_matrix(y_test, clf.predict(X_test))\n",
    "        \n",
    "            cv_errors[index] = accuracy(cv_conf)\n",
    "            train_errors[index] = accuracy(train_conf)\n",
    "\n",
    "        ax = axes[row, column]\n",
    "        ax.set_title('Gamma = %g'%gamma_value)\n",
    "        ax.semilogx(C_range, cv_errors, label='CV error')\n",
    "        ax.semilogx(C_range, train_errors, label='Train error')\n",
    "        plot_number += 1\n",
    "        ax.set_ylim([0.2,1])\n",
    "        \n",
    "    ax.legend(bbox_to_anchor=(1.05, 0), loc='lower left', borderaxespad=0.)\n",
    "    fig.text(0.5, 0.03, 'C value', ha='center',\n",
    "             fontsize=14)\n",
    "             \n",
    "    fig.text(0.04, 0.5, 'Classification Accuracy', va='center', \n",
    "             rotation='vertical', fontsize=14)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The best accuracy on the cross validation error curve was achieved for `gamma = 1`, and `C = 10`.  We can now create and train an optimized classifier based on these parameters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimized facies classification accuracy = 0.51\n",
      "Optimized adjacent facies classification accuracy = 0.88\n"
     ]
    }
   ],
   "source": [
    "clf = svm.LinearSVC(class_weight='balanced', tol=1e-03, random_state=42)        \n",
    "clf.fit(X_train, y_train)\n",
    "\n",
    "cv_conf = confusion_matrix(y_test, clf.predict(X_test))\n",
    "\n",
    "print('Optimized facies classification accuracy = %.2f' % accuracy(cv_conf))\n",
    "print('Optimized adjacent facies classification accuracy = %.2f' % accuracy_adjacent(cv_conf, adjacent_facies))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Pred    SS  CSiS  FSiS  SiSh    MS    WS     D    PS    BS Total\n",
      "     True\n",
      "       SS    40    14     1                                        55\n",
      "     CSiS    35   113    48                                   1   197\n",
      "     FSiS     6    39    80     1                 1     2     2   131\n",
      "     SiSh                 1    33           6    11     2     1    54\n",
      "       MS     1     2     1    11     5    40     6     5     4    75\n",
      "       WS                      13     2    58    10    11    15   109\n",
      "        D                       2                19     3     2    26\n",
      "       PS           1     1     5     3    41    14    37    36   138\n",
      "       BS                       1           3           7    39    50\n",
      "\n",
      "Precision  0.49  0.67  0.61  0.50  0.50  0.39  0.31  0.55  0.39  0.54\n",
      "   Recall  0.73  0.57  0.61  0.61  0.07  0.53  0.73  0.27  0.78  0.51\n",
      "       F1  0.58  0.62  0.61  0.55  0.12  0.45  0.44  0.36  0.52  0.49\n"
     ]
    }
   ],
   "source": [
    "display_cm(cv_conf, facies_labels, \n",
    "           display_metrics=True, hide_zeros=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Pred    SS  CSiS  FSiS  SiSh    MS    WS     D    PS    BS Total\n",
      "     True\n",
      "       SS    54           1                                        55\n",
      "     CSiS         196                                         1   197\n",
      "     FSiS     6         119     1                 1     2     2   131\n",
      "     SiSh                 1    33           6    11     2     1    54\n",
      "       MS     1     2     1          56           6     5     4    75\n",
      "       WS                      13          81                15   109\n",
      "        D                       2                22           2    26\n",
      "       PS           1     1     5     3               128         138\n",
      "       BS                       1           3                46    50\n",
      "\n",
      "Precision  0.89  0.98  0.97  0.60  0.95  0.90  0.55  0.93  0.65  0.89\n",
      "   Recall  0.98  0.99  0.91  0.61  0.75  0.74  0.85  0.93  0.92  0.88\n",
      "       F1  0.93  0.99  0.94  0.61  0.84  0.81  0.67  0.93  0.76  0.88\n"
     ]
    }
   ],
   "source": [
    "display_adj_cm(cv_conf, facies_labels, adjacent_facies, \n",
    "           display_metrics=True, hide_zeros=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Applying the classification model to the blind data\n",
    "\n",
    "We held a well back from the training, and stored it in a dataframe called `blind`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>Facies</th>\n",
       "      <th>Formation</th>\n",
       "      <th>Well Name</th>\n",
       "      <th>Depth</th>\n",
       "      <th>GR</th>\n",
       "      <th>ILD_log10</th>\n",
       "      <th>DeltaPHI</th>\n",
       "      <th>PHIND</th>\n",
       "      <th>PE</th>\n",
       "      <th>NM_M</th>\n",
       "      <th>RELPOS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>937</th>\n",
       "      <td>2</td>\n",
       "      <td>A1 SH</td>\n",
       "      <td>SHANKLE</td>\n",
       "      <td>2774.5</td>\n",
       "      <td>98.36</td>\n",
       "      <td>0.642</td>\n",
       "      <td>-0.1</td>\n",
       "      <td>18.685</td>\n",
       "      <td>2.9</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>938</th>\n",
       "      <td>2</td>\n",
       "      <td>A1 SH</td>\n",
       "      <td>SHANKLE</td>\n",
       "      <td>2775.0</td>\n",
       "      <td>97.57</td>\n",
       "      <td>0.631</td>\n",
       "      <td>7.9</td>\n",
       "      <td>16.745</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>939</th>\n",
       "      <td>2</td>\n",
       "      <td>A1 SH</td>\n",
       "      <td>SHANKLE</td>\n",
       "      <td>2775.5</td>\n",
       "      <td>98.41</td>\n",
       "      <td>0.615</td>\n",
       "      <td>12.8</td>\n",
       "      <td>14.105</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.968</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>940</th>\n",
       "      <td>2</td>\n",
       "      <td>A1 SH</td>\n",
       "      <td>SHANKLE</td>\n",
       "      <td>2776.0</td>\n",
       "      <td>85.92</td>\n",
       "      <td>0.597</td>\n",
       "      <td>13.0</td>\n",
       "      <td>13.385</td>\n",
       "      <td>3.4</td>\n",
       "      <td>1</td>\n",
       "      <td>0.952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>941</th>\n",
       "      <td>2</td>\n",
       "      <td>A1 SH</td>\n",
       "      <td>SHANKLE</td>\n",
       "      <td>2776.5</td>\n",
       "      <td>83.16</td>\n",
       "      <td>0.592</td>\n",
       "      <td>12.3</td>\n",
       "      <td>13.345</td>\n",
       "      <td>3.4</td>\n",
       "      <td>1</td>\n",
       "      <td>0.935</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1381</th>\n",
       "      <td>8</td>\n",
       "      <td>C LM</td>\n",
       "      <td>SHANKLE</td>\n",
       "      <td>3006.0</td>\n",
       "      <td>32.84</td>\n",
       "      <td>1.120</td>\n",
       "      <td>-2.2</td>\n",
       "      <td>3.455</td>\n",
       "      <td>5.1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.060</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1382</th>\n",
       "      <td>8</td>\n",
       "      <td>C LM</td>\n",
       "      <td>SHANKLE</td>\n",
       "      <td>3006.5</td>\n",
       "      <td>25.16</td>\n",
       "      <td>1.112</td>\n",
       "      <td>-1.6</td>\n",
       "      <td>2.890</td>\n",
       "      <td>4.8</td>\n",
       "      <td>2</td>\n",
       "      <td>0.045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1383</th>\n",
       "      <td>8</td>\n",
       "      <td>C LM</td>\n",
       "      <td>SHANKLE</td>\n",
       "      <td>3007.0</td>\n",
       "      <td>25.16</td>\n",
       "      <td>1.112</td>\n",
       "      <td>-1.6</td>\n",
       "      <td>2.890</td>\n",
       "      <td>4.8</td>\n",
       "      <td>2</td>\n",
       "      <td>0.030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1384</th>\n",
       "      <td>4</td>\n",
       "      <td>C LM</td>\n",
       "      <td>SHANKLE</td>\n",
       "      <td>3007.5</td>\n",
       "      <td>26.22</td>\n",
       "      <td>1.092</td>\n",
       "      <td>-0.4</td>\n",
       "      <td>3.400</td>\n",
       "      <td>4.5</td>\n",
       "      <td>2</td>\n",
       "      <td>0.030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1385</th>\n",
       "      <td>4</td>\n",
       "      <td>C LM</td>\n",
       "      <td>SHANKLE</td>\n",
       "      <td>3008.0</td>\n",
       "      <td>65.36</td>\n",
       "      <td>1.026</td>\n",
       "      <td>1.6</td>\n",
       "      <td>4.715</td>\n",
       "      <td>4.5</td>\n",
       "      <td>2</td>\n",
       "      <td>0.015</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>449 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Facies Formation Well Name   Depth     GR  ILD_log10  DeltaPHI   PHIND  \\\n",
       "937        2     A1 SH   SHANKLE  2774.5  98.36      0.642      -0.1  18.685   \n",
       "938        2     A1 SH   SHANKLE  2775.0  97.57      0.631       7.9  16.745   \n",
       "939        2     A1 SH   SHANKLE  2775.5  98.41      0.615      12.8  14.105   \n",
       "940        2     A1 SH   SHANKLE  2776.0  85.92      0.597      13.0  13.385   \n",
       "941        2     A1 SH   SHANKLE  2776.5  83.16      0.592      12.3  13.345   \n",
       "...      ...       ...       ...     ...    ...        ...       ...     ...   \n",
       "1381       8      C LM   SHANKLE  3006.0  32.84      1.120      -2.2   3.455   \n",
       "1382       8      C LM   SHANKLE  3006.5  25.16      1.112      -1.6   2.890   \n",
       "1383       8      C LM   SHANKLE  3007.0  25.16      1.112      -1.6   2.890   \n",
       "1384       4      C LM   SHANKLE  3007.5  26.22      1.092      -0.4   3.400   \n",
       "1385       4      C LM   SHANKLE  3008.0  65.36      1.026       1.6   4.715   \n",
       "\n",
       "       PE  NM_M  RELPOS  \n",
       "937   2.9     1   1.000  \n",
       "938   3.2     1   0.984  \n",
       "939   3.2     1   0.968  \n",
       "940   3.4     1   0.952  \n",
       "941   3.4     1   0.935  \n",
       "...   ...   ...     ...  \n",
       "1381  5.1     2   0.060  \n",
       "1382  4.8     2   0.045  \n",
       "1383  4.8     2   0.030  \n",
       "1384  4.5     2   0.030  \n",
       "1385  4.5     2   0.015  \n",
       "\n",
       "[449 rows x 11 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "blind"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The label vector is just the `Facies` column:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y_blind = blind['Facies'].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can form the feature matrix by dropping some of the columns and making a new dataframe:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "well_features = blind.drop(['Facies', 'Formation', 'Well Name', 'Depth'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can transform this with the scaler we made before:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "X_blind = scaler.transform(well_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now it's a simple matter of making a prediction and storing it back in the dataframe:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "y_pred = clf.predict(X_blind)\n",
    "blind['Prediction'] = y_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see how we did with the confusion matrix:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimized facies classification accuracy = 0.55\n",
      "Optimized adjacent facies classification accuracy = 0.93\n"
     ]
    }
   ],
   "source": [
    "cv_conf = confusion_matrix(y_blind, y_pred)\n",
    "\n",
    "print('Optimized facies classification accuracy = %.2f' % accuracy(cv_conf))\n",
    "print('Optimized adjacent facies classification accuracy = %.2f' % accuracy_adjacent(cv_conf, adjacent_facies))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We managed 0.75 using the test data, but it was from the same wells as the training data. This more reasonable test does not perform as well..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Pred    SS  CSiS  FSiS  SiSh    MS    WS     D    PS    BS Total\n",
      "     True\n",
      "       SS    75    14                                              89\n",
      "     CSiS    28    43    16     2                                  89\n",
      "     FSiS     4    61    52                                       117\n",
      "     SiSh                       5           2                       7\n",
      "       MS                      12     2     2     2     1          19\n",
      "       WS                      14     1    45     3     7     1    71\n",
      "        D                                        17                17\n",
      "       PS                       1     5    17           9     8    40\n",
      "       BS                                                           0\n",
      "\n",
      "Precision  0.70  0.36  0.76  0.15  0.25  0.68  0.77  0.53  0.00  0.61\n",
      "   Recall  0.84  0.48  0.44  0.71  0.11  0.63  1.00  0.23  0.00  0.55\n",
      "       F1  0.77  0.42  0.56  0.24  0.15  0.66  0.87  0.32  0.00  0.56\n"
     ]
    }
   ],
   "source": [
    "display_cm(cv_conf, facies_labels,\n",
    "           display_metrics=True, hide_zeros=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "...but does remarkably well on the adjacent facies predictions. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Pred    SS  CSiS  FSiS  SiSh    MS    WS     D    PS    BS Total\n",
      "     True\n",
      "       SS    89                                                    89\n",
      "     CSiS          87           2                                  89\n",
      "     FSiS     4         113                                       117\n",
      "     SiSh                       5           2                       7\n",
      "       MS                            16           2     1          19\n",
      "       WS                      14          56                 1    71\n",
      "        D                                        17                17\n",
      "       PS                       1     5                34          40\n",
      "       BS                                                           0\n",
      "\n",
      "Precision  0.96  1.00  1.00  0.23  0.76  0.97  0.89  0.97  0.00  0.96\n",
      "   Recall  1.00  0.98  0.97  0.71  0.84  0.79  1.00  0.85  0.00  0.93\n",
      "       F1  0.98  0.99  0.98  0.34  0.80  0.87  0.94  0.91  0.00  0.94\n"
     ]
    }
   ],
   "source": [
    "display_adj_cm(cv_conf, facies_labels, adjacent_facies,\n",
    "               display_metrics=True, hide_zeros=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def compare_facies_plot(logs, compadre, facies_colors):\n",
    "    #make sure logs are sorted by depth\n",
    "    logs = logs.sort_values(by='Depth')\n",
    "    cmap_facies = colors.ListedColormap(\n",
    "            facies_colors[0:len(facies_colors)], 'indexed')\n",
    "    \n",
    "    ztop=logs.Depth.min(); zbot=logs.Depth.max()\n",
    "    \n",
    "    cluster1 = np.repeat(np.expand_dims(logs['Facies'].values,1), 100, 1)\n",
    "    cluster2 = np.repeat(np.expand_dims(logs[compadre].values,1), 100, 1)\n",
    "    \n",
    "    f, ax = plt.subplots(nrows=1, ncols=7, figsize=(9, 12))\n",
    "    ax[0].plot(logs.GR, logs.Depth, '-g')\n",
    "    ax[1].plot(logs.ILD_log10, logs.Depth, '-')\n",
    "    ax[2].plot(logs.DeltaPHI, logs.Depth, '-', color='0.5')\n",
    "    ax[3].plot(logs.PHIND, logs.Depth, '-', color='r')\n",
    "    ax[4].plot(logs.PE, logs.Depth, '-', color='black')\n",
    "    im1 = ax[5].imshow(cluster1, interpolation='none', aspect='auto',\n",
    "                    cmap=cmap_facies,vmin=1,vmax=9)\n",
    "    im2 = ax[6].imshow(cluster2, interpolation='none', aspect='auto',\n",
    "                    cmap=cmap_facies,vmin=1,vmax=9)\n",
    "    \n",
    "    divider = make_axes_locatable(ax[6])\n",
    "    cax = divider.append_axes(\"right\", size=\"20%\", pad=0.05)\n",
    "    cbar=plt.colorbar(im2, cax=cax)\n",
    "    cbar.set_label((17*' ').join([' SS ', 'CSiS', 'FSiS', \n",
    "                                'SiSh', ' MS ', ' WS ', ' D  ', \n",
    "                                ' PS ', ' BS ']))\n",
    "    cbar.set_ticks(range(0,1)); cbar.set_ticklabels('')\n",
    "    \n",
    "    for i in range(len(ax)-2):\n",
    "        ax[i].set_ylim(ztop,zbot)\n",
    "        ax[i].invert_yaxis()\n",
    "        ax[i].grid()\n",
    "        ax[i].locator_params(axis='x', nbins=3)\n",
    "    \n",
    "    ax[0].set_xlabel(\"GR\")\n",
    "    ax[0].set_xlim(logs.GR.min(),logs.GR.max())\n",
    "    ax[1].set_xlabel(\"ILD_log10\")\n",
    "    ax[1].set_xlim(logs.ILD_log10.min(),logs.ILD_log10.max())\n",
    "    ax[2].set_xlabel(\"DeltaPHI\")\n",
    "    ax[2].set_xlim(logs.DeltaPHI.min(),logs.DeltaPHI.max())\n",
    "    ax[3].set_xlabel(\"PHIND\")\n",
    "    ax[3].set_xlim(logs.PHIND.min(),logs.PHIND.max())\n",
    "    ax[4].set_xlabel(\"PE\")\n",
    "    ax[4].set_xlim(logs.PE.min(),logs.PE.max())\n",
    "    ax[5].set_xlabel('Facies')\n",
    "    ax[6].set_xlabel(compadre)\n",
    "    \n",
    "    ax[1].set_yticklabels([]); ax[2].set_yticklabels([]); ax[3].set_yticklabels([])\n",
    "    ax[4].set_yticklabels([]); ax[5].set_yticklabels([])\n",
    "    ax[5].set_xticklabels([])\n",
    "    ax[6].set_xticklabels([])\n",
    "    f.suptitle('Well: %s'%logs.iloc[0]['Well Name'], fontsize=14,y=0.94)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjAAAALsCAYAAAD9DY07AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUCi8jv8PVQxf/VKwVJkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1146014e0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "compare_facies_plot(blind, 'Prediction', facies_colors)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## K-fold cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.cross_validation import KFold\n",
    "kf = KFold(4, n_folds=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sklearn.cross_validation.KFold(n=4, n_folds=2, shuffle=False, random_state=None)\n"
     ]
    }
   ],
   "source": [
    "print(kf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "for train_index, test_index in kf:\n",
    "    print(\"TRAIN:\", train_index, \"TEST:\", test_index)\n",
    "    X_train, X_test = X[train_index], X[test_index]\n",
    "    y_train, y_test = y[train_index], y[test_index]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Applying the classification model to new data\n",
    "\n",
    "Now that we have a trained facies classification model we can use it to identify facies in wells that do not have core data.  In this case, we will apply the classifier to two wells, but we could use it on any number of wells for which we have the same set of well logs for input.\n",
    "\n",
    "This dataset is similar to the training data except it does not have facies labels.  It is loaded into a dataframe called `test_data`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "well_data = pd.read_csv('../validation_data_nofacies.csv')\n",
    "well_data['Well Name'] = well_data['Well Name'].astype('category')\n",
    "well_features = well_data.drop(['Formation', 'Well Name', 'Depth'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data needs to be scaled using the same constants we used for the training data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_unknown = scaler.transform(well_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally we predict facies labels for the unknown data, and store the results in a `Facies` column of the `test_data` dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>Formation</th>\n",
       "      <th>Well Name</th>\n",
       "      <th>Depth</th>\n",
       "      <th>GR</th>\n",
       "      <th>ILD_log10</th>\n",
       "      <th>DeltaPHI</th>\n",
       "      <th>PHIND</th>\n",
       "      <th>PE</th>\n",
       "      <th>NM_M</th>\n",
       "      <th>RELPOS</th>\n",
       "      <th>Facies</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A1 SH</td>\n",
       "      <td>STUART</td>\n",
       "      <td>2808.0</td>\n",
       "      <td>66.276</td>\n",
       "      <td>0.630</td>\n",
       "      <td>3.300</td>\n",
       "      <td>10.650</td>\n",
       "      <td>3.591</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A1 SH</td>\n",
       "      <td>STUART</td>\n",
       "      <td>2808.5</td>\n",
       "      <td>77.252</td>\n",
       "      <td>0.585</td>\n",
       "      <td>6.500</td>\n",
       "      <td>11.950</td>\n",
       "      <td>3.341</td>\n",
       "      <td>1</td>\n",
       "      <td>0.978</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A1 SH</td>\n",
       "      <td>STUART</td>\n",
       "      <td>2809.0</td>\n",
       "      <td>82.899</td>\n",
       "      <td>0.566</td>\n",
       "      <td>9.400</td>\n",
       "      <td>13.600</td>\n",
       "      <td>3.064</td>\n",
       "      <td>1</td>\n",
       "      <td>0.956</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>A1 SH</td>\n",
       "      <td>STUART</td>\n",
       "      <td>2809.5</td>\n",
       "      <td>80.671</td>\n",
       "      <td>0.593</td>\n",
       "      <td>9.500</td>\n",
       "      <td>13.250</td>\n",
       "      <td>2.977</td>\n",
       "      <td>1</td>\n",
       "      <td>0.933</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>A1 SH</td>\n",
       "      <td>STUART</td>\n",
       "      <td>2810.0</td>\n",
       "      <td>75.971</td>\n",
       "      <td>0.638</td>\n",
       "      <td>8.700</td>\n",
       "      <td>12.350</td>\n",
       "      <td>3.020</td>\n",
       "      <td>1</td>\n",
       "      <td>0.911</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>825</th>\n",
       "      <td>C SH</td>\n",
       "      <td>CRAWFORD</td>\n",
       "      <td>3158.5</td>\n",
       "      <td>86.078</td>\n",
       "      <td>0.554</td>\n",
       "      <td>5.040</td>\n",
       "      <td>16.150</td>\n",
       "      <td>3.161</td>\n",
       "      <td>1</td>\n",
       "      <td>0.639</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>826</th>\n",
       "      <td>C SH</td>\n",
       "      <td>CRAWFORD</td>\n",
       "      <td>3159.0</td>\n",
       "      <td>88.855</td>\n",
       "      <td>0.539</td>\n",
       "      <td>5.560</td>\n",
       "      <td>16.750</td>\n",
       "      <td>3.118</td>\n",
       "      <td>1</td>\n",
       "      <td>0.611</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>827</th>\n",
       "      <td>C SH</td>\n",
       "      <td>CRAWFORD</td>\n",
       "      <td>3159.5</td>\n",
       "      <td>90.490</td>\n",
       "      <td>0.530</td>\n",
       "      <td>6.360</td>\n",
       "      <td>16.780</td>\n",
       "      <td>3.168</td>\n",
       "      <td>1</td>\n",
       "      <td>0.583</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>828</th>\n",
       "      <td>C SH</td>\n",
       "      <td>CRAWFORD</td>\n",
       "      <td>3160.0</td>\n",
       "      <td>90.975</td>\n",
       "      <td>0.522</td>\n",
       "      <td>7.035</td>\n",
       "      <td>16.995</td>\n",
       "      <td>3.154</td>\n",
       "      <td>1</td>\n",
       "      <td>0.556</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>829</th>\n",
       "      <td>C SH</td>\n",
       "      <td>CRAWFORD</td>\n",
       "      <td>3160.5</td>\n",
       "      <td>90.108</td>\n",
       "      <td>0.513</td>\n",
       "      <td>7.505</td>\n",
       "      <td>17.595</td>\n",
       "      <td>3.125</td>\n",
       "      <td>1</td>\n",
       "      <td>0.528</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>830 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Formation Well Name   Depth      GR  ILD_log10  DeltaPHI   PHIND     PE  \\\n",
       "0       A1 SH    STUART  2808.0  66.276      0.630     3.300  10.650  3.591   \n",
       "1       A1 SH    STUART  2808.5  77.252      0.585     6.500  11.950  3.341   \n",
       "2       A1 SH    STUART  2809.0  82.899      0.566     9.400  13.600  3.064   \n",
       "3       A1 SH    STUART  2809.5  80.671      0.593     9.500  13.250  2.977   \n",
       "4       A1 SH    STUART  2810.0  75.971      0.638     8.700  12.350  3.020   \n",
       "..        ...       ...     ...     ...        ...       ...     ...    ...   \n",
       "825      C SH  CRAWFORD  3158.5  86.078      0.554     5.040  16.150  3.161   \n",
       "826      C SH  CRAWFORD  3159.0  88.855      0.539     5.560  16.750  3.118   \n",
       "827      C SH  CRAWFORD  3159.5  90.490      0.530     6.360  16.780  3.168   \n",
       "828      C SH  CRAWFORD  3160.0  90.975      0.522     7.035  16.995  3.154   \n",
       "829      C SH  CRAWFORD  3160.5  90.108      0.513     7.505  17.595  3.125   \n",
       "\n",
       "     NM_M  RELPOS  Facies  \n",
       "0       1   1.000       2  \n",
       "1       1   0.978       3  \n",
       "2       1   0.956       3  \n",
       "3       1   0.933       2  \n",
       "4       1   0.911       2  \n",
       "..    ...     ...     ...  \n",
       "825     1   0.639       2  \n",
       "826     1   0.611       2  \n",
       "827     1   0.583       3  \n",
       "828     1   0.556       3  \n",
       "829     1   0.528       3  \n",
       "\n",
       "[830 rows x 11 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#predict facies of unclassified data\n",
    "y_unknown = clf.predict(X_unknown)\n",
    "well_data['Facies'] = y_unknown\n",
    "well_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[STUART, CRAWFORD]\n",
       "Categories (2, object): [STUART, CRAWFORD]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "well_data['Well Name'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can use the well log plot to view the classification results along with the well logs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "well_data.to_csv('well_data_with_facies.csv')"
   ]
  },
  {
   "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
}