Kr1m/Kr1m_SEG_ML_Attempt1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Facies classification utilizing an Adaptive Boosted Random Forest\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[Ryan Thielke](http://www.linkedin.com/in/ryan-thielke-b987012a)\n",
    "\n",
    "\n",
    "In the following, we provide a possible solution to the facies classification problem described in https://github.com/seg/2016-ml-contest.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploring the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "%matplotlib inline\n",
    "\n",
    "import sys\n",
    "sys.path.append(\"..\")\n",
    "\n",
    "#Import standard pydata libs\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>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>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>3232.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.503254</td>\n",
       "      <td>2906.867438</td>\n",
       "      <td>64.933985</td>\n",
       "      <td>0.659566</td>\n",
       "      <td>4.402484</td>\n",
       "      <td>13.201066</td>\n",
       "      <td>3.725014</td>\n",
       "      <td>1.518438</td>\n",
       "      <td>0.521852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.474324</td>\n",
       "      <td>133.300164</td>\n",
       "      <td>30.302530</td>\n",
       "      <td>0.252703</td>\n",
       "      <td>5.274947</td>\n",
       "      <td>7.132846</td>\n",
       "      <td>0.896152</td>\n",
       "      <td>0.499720</td>\n",
       "      <td>0.286644</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>2821.500000</td>\n",
       "      <td>44.730000</td>\n",
       "      <td>0.498000</td>\n",
       "      <td>1.600000</td>\n",
       "      <td>8.500000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.277000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>2932.500000</td>\n",
       "      <td>64.990000</td>\n",
       "      <td>0.639000</td>\n",
       "      <td>4.300000</td>\n",
       "      <td>12.020000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.528000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>3007.000000</td>\n",
       "      <td>79.438000</td>\n",
       "      <td>0.822000</td>\n",
       "      <td>7.500000</td>\n",
       "      <td>16.050000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.769000</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  4149.000000  4149.000000  4149.000000  4149.000000  4149.000000   \n",
       "mean      4.503254  2906.867438    64.933985     0.659566     4.402484   \n",
       "std       2.474324   133.300164    30.302530     0.252703     5.274947   \n",
       "min       1.000000  2573.500000    10.149000    -0.025949   -21.832000   \n",
       "25%       2.000000  2821.500000    44.730000     0.498000     1.600000   \n",
       "50%       4.000000  2932.500000    64.990000     0.639000     4.300000   \n",
       "75%       6.000000  3007.000000    79.438000     0.822000     7.500000   \n",
       "max       9.000000  3138.000000   361.150000     1.800000    19.312000   \n",
       "\n",
       "             PHIND           PE         NM_M       RELPOS  \n",
       "count  4149.000000  3232.000000  4149.000000  4149.000000  \n",
       "mean     13.201066     3.725014     1.518438     0.521852  \n",
       "std       7.132846     0.896152     0.499720     0.286644  \n",
       "min       0.550000     0.200000     1.000000     0.000000  \n",
       "25%       8.500000          NaN     1.000000     0.277000  \n",
       "50%      12.020000          NaN     2.000000     0.528000  \n",
       "75%      16.050000          NaN     2.000000     0.769000  \n",
       "max      84.400000     8.094000     2.000000     1.000000  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filename = '../facies_vectors.csv'\n",
    "training_data = pd.read_csv(filename)\n",
    "training_data['Well Name'] = training_data['Well Name'].astype('category')\n",
    "training_data['Formation'] = training_data['Formation'].astype('category')                            \n",
    "training_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJEAAASHCAYAAABSwew0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngKH/ByzW+wWWORMmAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10a43b278>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Visualize the distribution of facies for each well\n",
    "wells = training_data['Well Name'].unique()\n",
    "\n",
    "fig, ax = plt.subplots(5,2, figsize=(20,20))\n",
    "for i, well in enumerate(wells):\n",
    "    row = i % ax.shape[0]\n",
    "    column = i // ax.shape[0]\n",
    "    counts = training_data[training_data['Well Name']==well].Facies.value_counts()\n",
    "    data_for_well = [counts[j] if j in counts.index else 0 for j in range(1,10)]\n",
    "    ax[row, column].bar(range(1,10), data_for_well, align='center')\n",
    "    ax[row, column].set_title(\"{well}\".format(well=well))\n",
    "    ax[row, column].set_ylabel(\"Counts\")\n",
    "    ax[row, column].set_xticks(range(1,10))\n",
    "\n",
    "plt.show()\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10ba542e8>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAi8AAAJMCAYAAAAytEawAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nViheAACArVC8AAAAW/n/AzQ+p/bPjJsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10a43bac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "sns.heatmap(training_data.drop(['Formation', 'Well Name'], axis=1).corr())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature Engineering\n",
    "\n",
    "Here we will do a couple things to clean the data and attempt to create new features for our model to consume.\n",
    "\n",
    "First, we will smooth the PE and GR features.\n",
    "Second, we replace missing PE values with the mean of the entire dataset (might want to investigate other methods)\n",
    "Last, we will encode the formations into integer values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dfs = []\n",
    "for well in training_data['Well Name'].unique():\n",
    "    df = training_data[training_data['Well Name']==well].copy(deep=True)\n",
    "    df.sort_values('Depth', inplace=True)\n",
    "    for col in ['PE', 'GR']:\n",
    "        smooth_col = 'smooth_'+col\n",
    "        df[smooth_col] = pd.rolling_mean(df[col], window=25)\n",
    "        df[smooth_col].fillna(method='ffill', inplace=True)\n",
    "        df[smooth_col].fillna(method='bfill', inplace=True)\n",
    "    dfs.append(df)\n",
    "training_data = pd.concat(dfs)\n",
    "pe_mean = training_data.PE.mean()\n",
    "sm_pe_mean = training_data.smooth_PE.mean()\n",
    "training_data['PE'] = training_data.PE.replace({np.nan:pe_mean})\n",
    "training_data['smooth_PE'] = training_data['smooth_PE'].replace({np.nan:sm_pe_mean})\n",
    "formation_encoder = dict(zip(training_data.Formation.unique(), range(len(training_data.Formation.unique()))))\n",
    "training_data['enc_formation'] = training_data.Formation.map(formation_encoder)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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",
       "      <th>smooth_PE</th>\n",
       "      <th>smooth_GR</th>\n",
       "      <th>enc_formation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "      <td>4149.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.503254</td>\n",
       "      <td>2906.867438</td>\n",
       "      <td>64.933985</td>\n",
       "      <td>0.659566</td>\n",
       "      <td>4.402484</td>\n",
       "      <td>13.201066</td>\n",
       "      <td>3.725014</td>\n",
       "      <td>1.518438</td>\n",
       "      <td>0.521852</td>\n",
       "      <td>3.707428</td>\n",
       "      <td>65.006645</td>\n",
       "      <td>6.476982</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.474324</td>\n",
       "      <td>133.300164</td>\n",
       "      <td>30.302530</td>\n",
       "      <td>0.252703</td>\n",
       "      <td>5.274947</td>\n",
       "      <td>7.132846</td>\n",
       "      <td>0.790917</td>\n",
       "      <td>0.499720</td>\n",
       "      <td>0.286644</td>\n",
       "      <td>0.577399</td>\n",
       "      <td>17.544534</td>\n",
       "      <td>4.819176</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",
       "      <td>2.552000</td>\n",
       "      <td>22.672680</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>2821.500000</td>\n",
       "      <td>44.730000</td>\n",
       "      <td>0.498000</td>\n",
       "      <td>1.600000</td>\n",
       "      <td>8.500000</td>\n",
       "      <td>3.200000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.277000</td>\n",
       "      <td>3.329280</td>\n",
       "      <td>53.656400</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>2932.500000</td>\n",
       "      <td>64.990000</td>\n",
       "      <td>0.639000</td>\n",
       "      <td>4.300000</td>\n",
       "      <td>12.020000</td>\n",
       "      <td>3.725014</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.528000</td>\n",
       "      <td>3.707428</td>\n",
       "      <td>65.104800</td>\n",
       "      <td>6.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>3007.000000</td>\n",
       "      <td>79.438000</td>\n",
       "      <td>0.822000</td>\n",
       "      <td>7.500000</td>\n",
       "      <td>16.050000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.769000</td>\n",
       "      <td>3.916000</td>\n",
       "      <td>75.716000</td>\n",
       "      <td>12.000000</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",
       "      <td>6.086440</td>\n",
       "      <td>134.212600</td>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Facies        Depth           GR    ILD_log10     DeltaPHI  \\\n",
       "count  4149.000000  4149.000000  4149.000000  4149.000000  4149.000000   \n",
       "mean      4.503254  2906.867438    64.933985     0.659566     4.402484   \n",
       "std       2.474324   133.300164    30.302530     0.252703     5.274947   \n",
       "min       1.000000  2573.500000    10.149000    -0.025949   -21.832000   \n",
       "25%       2.000000  2821.500000    44.730000     0.498000     1.600000   \n",
       "50%       4.000000  2932.500000    64.990000     0.639000     4.300000   \n",
       "75%       6.000000  3007.000000    79.438000     0.822000     7.500000   \n",
       "max       9.000000  3138.000000   361.150000     1.800000    19.312000   \n",
       "\n",
       "             PHIND           PE         NM_M       RELPOS    smooth_PE  \\\n",
       "count  4149.000000  4149.000000  4149.000000  4149.000000  4149.000000   \n",
       "mean     13.201066     3.725014     1.518438     0.521852     3.707428   \n",
       "std       7.132846     0.790917     0.499720     0.286644     0.577399   \n",
       "min       0.550000     0.200000     1.000000     0.000000     2.552000   \n",
       "25%       8.500000     3.200000     1.000000     0.277000     3.329280   \n",
       "50%      12.020000     3.725014     2.000000     0.528000     3.707428   \n",
       "75%      16.050000     4.000000     2.000000     0.769000     3.916000   \n",
       "max      84.400000     8.094000     2.000000     1.000000     6.086440   \n",
       "\n",
       "         smooth_GR  enc_formation  \n",
       "count  4149.000000    4149.000000  \n",
       "mean     65.006645       6.476982  \n",
       "std      17.544534       4.819176  \n",
       "min      22.672680       0.000000  \n",
       "25%      53.656400       2.000000  \n",
       "50%      65.104800       6.000000  \n",
       "75%      75.716000      12.000000  \n",
       "max     134.212600      13.000000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building the model and parameter tuning\n",
    "\n",
    "In the section below we will create a Adaptive Boosted Random Forest Classifier from the Scikit-Learn ML Library"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#Let's build a model\n",
    "from sklearn import preprocessing\n",
    "from sklearn.ensemble import AdaBoostClassifier, RandomForestClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn import metrics, cross_validation \n",
    "from classification_utilities import display_cm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "********************\n",
      "SHRIMPLIN=0.6306\n",
      "********************\n",
      "********************\n",
      "ALEXANDER D=0.6738\n",
      "********************\n",
      "********************\n",
      "SHANKLE=0.4477\n",
      "********************\n",
      "********************\n",
      "LUKE G U=0.6334\n",
      "********************\n",
      "********************\n",
      "KIMZEY A=0.5148\n",
      "********************\n",
      "********************\n",
      "CROSS H CATTLE=0.4331\n",
      "********************\n",
      "********************\n",
      "NOLAN=0.4892\n",
      "********************\n",
      "********************\n",
      "Recruit F9=0.6750\n",
      "********************\n",
      "********************\n",
      "NEWBY=0.4320\n",
      "********************\n",
      "********************\n",
      "CHURCHMAN BIBLE=0.5248\n",
      "********************\n"
     ]
    }
   ],
   "source": [
    "#We will take a look at an F1 score for each well\n",
    "n_estimators=100\n",
    "learning_rate=.01\n",
    "random_state=0\n",
    "facies_labels = ['SS', 'CSiS', 'FSiS', 'SiSh', 'MS',\n",
    "                 'WS', 'D','PS', 'BS']\n",
    "\n",
    "title_length = 20 \n",
    "\n",
    "wells = training_data['Well Name'].unique()\n",
    "for well in wells:\n",
    "    blind = training_data[training_data['Well Name']==well]\n",
    "    train = training_data[(training_data['Well Name']!=well)]\n",
    "    \n",
    "    train_X = train.drop(['Formation', 'Well Name', 'Depth', 'Facies'], axis=1)\n",
    "    train_Y = train.Facies.values\n",
    "    test_X = blind.drop(['Formation', 'Well Name', 'Facies', 'Depth'], axis=1)\n",
    "    test_Y = blind.Facies.values\n",
    "    \n",
    "    clf = AdaBoostClassifier(RandomForestClassifier(), n_estimators=200, learning_rate=learning_rate, random_state=random_state, algorithm='SAMME.R')\n",
    "    clf.fit(X=train_X, y=train_Y)\n",
    "    pred_Y = clf.predict(test_X)\n",
    "    f1 = metrics.f1_score(test_Y, pred_Y, average='micro')\n",
    "    print(\"*\"*title_length)\n",
    "    print(\"{well}={f1:.4f}\".format(well=well,f1=f1))\n",
    "    print(\"*\"*title_length)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "train_X, test_X, train_Y, test_Y = cross_validation.train_test_split(training_data.drop(['Formation', 'Well Name','Facies', 'Depth'], axis=1), training_data.Facies.values, test_size=.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3319, 10)\n",
      "(3319,)\n",
      "(830, 10)\n",
      "(830,)\n"
     ]
    }
   ],
   "source": [
    "print(train_X.shape)\n",
    "print(train_Y.shape)\n",
    "print(test_X.shape)\n",
    "print(test_Y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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    50    16     1     0     0     0     0     0     0    67\n",
      "     CSiS     2   164    25     0     0     0     0     1     0   192\n",
      "     FSiS     0    18   134     0     0     2     0     1     0   155\n",
      "     SiSh     0     0     1    53     0     4     0     2     0    60\n",
      "       MS     0     2     2     4    31    11     1     9     0    60\n",
      "       WS     0     0     2     5     3    81     1    19     1   112\n",
      "        D     0     0     1     0     1     0    22     6     0    30\n",
      "       PS     0     0     3     1     1     9     0    98     0   112\n",
      "       BS     0     0     0     0     0     1     0     2    39    42\n",
      "\n",
      "Precision  0.96  0.82  0.79  0.84  0.86  0.75  0.92  0.71  0.97  0.82\n",
      "   Recall  0.75  0.85  0.86  0.88  0.52  0.72  0.73  0.88  0.93  0.81\n",
      "       F1  0.84  0.84  0.83  0.86  0.65  0.74  0.81  0.78  0.95  0.81\n"
     ]
    }
   ],
   "source": [
    "clf = AdaBoostClassifier(RandomForestClassifier(), n_estimators=200, learning_rate=learning_rate, random_state=0, algorithm='SAMME.R')\n",
    "\n",
    "clf.fit(train_X, train_Y)\n",
    "pred_Y = clf.predict(test_X)\n",
    "cm = metrics.confusion_matrix(y_true=test_Y, y_pred=pred_Y)\n",
    "display_cm(cm, facies_labels, display_metrics=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "validation_data = pd.read_csv(\"../validation_data_nofacies.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>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>smooth_PE</th>\n",
       "      <th>smooth_GR</th>\n",
       "      <th>enc_formation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>830.000000</td>\n",
       "      <td>830.00000</td>\n",
       "      <td>830.000000</td>\n",
       "      <td>830.000000</td>\n",
       "      <td>830.000000</td>\n",
       "      <td>830.000000</td>\n",
       "      <td>830.000000</td>\n",
       "      <td>830.000000</td>\n",
       "      <td>830.000000</td>\n",
       "      <td>830.000000</td>\n",
       "      <td>830.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2987.070482</td>\n",
       "      <td>57.61173</td>\n",
       "      <td>0.666312</td>\n",
       "      <td>2.851964</td>\n",
       "      <td>11.655277</td>\n",
       "      <td>3.654178</td>\n",
       "      <td>1.678313</td>\n",
       "      <td>0.535807</td>\n",
       "      <td>3.640066</td>\n",
       "      <td>57.535834</td>\n",
       "      <td>6.214458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>94.391925</td>\n",
       "      <td>27.52774</td>\n",
       "      <td>0.288367</td>\n",
       "      <td>3.442074</td>\n",
       "      <td>5.190236</td>\n",
       "      <td>0.649793</td>\n",
       "      <td>0.467405</td>\n",
       "      <td>0.283062</td>\n",
       "      <td>0.373720</td>\n",
       "      <td>16.020556</td>\n",
       "      <td>4.599762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2808.000000</td>\n",
       "      <td>12.03600</td>\n",
       "      <td>-0.468000</td>\n",
       "      <td>-8.900000</td>\n",
       "      <td>1.855000</td>\n",
       "      <td>2.113000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.013000</td>\n",
       "      <td>2.783080</td>\n",
       "      <td>23.462960</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2911.625000</td>\n",
       "      <td>36.77325</td>\n",
       "      <td>0.541000</td>\n",
       "      <td>0.411250</td>\n",
       "      <td>7.700000</td>\n",
       "      <td>3.171500</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>3.386970</td>\n",
       "      <td>46.641210</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2993.750000</td>\n",
       "      <td>58.34450</td>\n",
       "      <td>0.675000</td>\n",
       "      <td>2.397500</td>\n",
       "      <td>10.950000</td>\n",
       "      <td>3.515500</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.547500</td>\n",
       "      <td>3.631880</td>\n",
       "      <td>56.626940</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3055.375000</td>\n",
       "      <td>73.05150</td>\n",
       "      <td>0.850750</td>\n",
       "      <td>4.600000</td>\n",
       "      <td>14.793750</td>\n",
       "      <td>4.191500</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.778000</td>\n",
       "      <td>3.846550</td>\n",
       "      <td>68.118700</td>\n",
       "      <td>11.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3160.500000</td>\n",
       "      <td>220.41300</td>\n",
       "      <td>1.507000</td>\n",
       "      <td>16.500000</td>\n",
       "      <td>31.335000</td>\n",
       "      <td>6.321000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.696200</td>\n",
       "      <td>95.138680</td>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Depth         GR   ILD_log10    DeltaPHI       PHIND          PE  \\\n",
       "count   830.000000  830.00000  830.000000  830.000000  830.000000  830.000000   \n",
       "mean   2987.070482   57.61173    0.666312    2.851964   11.655277    3.654178   \n",
       "std      94.391925   27.52774    0.288367    3.442074    5.190236    0.649793   \n",
       "min    2808.000000   12.03600   -0.468000   -8.900000    1.855000    2.113000   \n",
       "25%    2911.625000   36.77325    0.541000    0.411250    7.700000    3.171500   \n",
       "50%    2993.750000   58.34450    0.675000    2.397500   10.950000    3.515500   \n",
       "75%    3055.375000   73.05150    0.850750    4.600000   14.793750    4.191500   \n",
       "max    3160.500000  220.41300    1.507000   16.500000   31.335000    6.321000   \n",
       "\n",
       "             NM_M      RELPOS   smooth_PE   smooth_GR  enc_formation  \n",
       "count  830.000000  830.000000  830.000000  830.000000     830.000000  \n",
       "mean     1.678313    0.535807    3.640066   57.535834       6.214458  \n",
       "std      0.467405    0.283062    0.373720   16.020556       4.599762  \n",
       "min      1.000000    0.013000    2.783080   23.462960       0.000000  \n",
       "25%      1.000000    0.300000    3.386970   46.641210       1.000000  \n",
       "50%      2.000000    0.547500    3.631880   56.626940       5.000000  \n",
       "75%      2.000000    0.778000    3.846550   68.118700      11.000000  \n",
       "max      2.000000    1.000000    4.696200   95.138680      13.000000  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfs = []\n",
    "for well in validation_data['Well Name'].unique():\n",
    "    df = validation_data[validation_data['Well Name']==well].copy(deep=True)\n",
    "    df.sort_values('Depth', inplace=True)\n",
    "    for col in ['PE', 'GR']:\n",
    "        smooth_col = 'smooth_'+col\n",
    "        df[smooth_col] = pd.rolling_mean(df[col], window=25)\n",
    "        df[smooth_col].fillna(method='ffill', inplace=True)\n",
    "        df[smooth_col].fillna(method='bfill', inplace=True)\n",
    "    dfs.append(df)\n",
    "validation_data = pd.concat(dfs)\n",
    "\n",
    "validation_data['enc_formation'] = validation_data.Formation.map(formation_encoder)\n",
    "validation_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "X = training_data.drop(['Formation', 'Well Name', 'Depth','Facies'], axis=1)\n",
    "Y = training_data.Facies.values\n",
    "test_X = validation_data.drop(['Formation', 'Well Name', 'Depth'], axis=1)\n",
    "\n",
    "clf = AdaBoostClassifier(RandomForestClassifier(), n_estimators=200, learning_rate=learning_rate, random_state=0)\n",
    "clf.fit(X,Y)\n",
    "predicted_facies = clf.predict(test_X)\n",
    "validation_data['Facies'] = predicted_facies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "validation_data.to_csv(\"Kr1m_SEG_ML_Attempt1.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}