SCIO-NCH_BENE_IP_DDCTBL_AMT_PREDICTOR.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import mpl_toolkits\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data = pd.read_csv(\"C://Users//WELCOME//Desktop//SCIO-images//NCH_BENE_IP_DDCTBL_AMT_PREDICTOR.csv\")\n",
    "data_no_nan = data.fillna(0)\n",
    "data_no_nan = data_no_nan.fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CLM_ID</th>\n",
       "      <th>CLM_PMT_AMT</th>\n",
       "      <th>CLM_PASS_THRU_PER_DIEM_AMT</th>\n",
       "      <th>CLM_UTLZTN_DAY_CNT</th>\n",
       "      <th>CLM_YEAR_MONTH_CLM_FROM_DT_CAT</th>\n",
       "      <th>CLM_YEAR_MONTH_CLM_THRU_DT_CAT</th>\n",
       "      <th>NCH_BENE_DSCHRG_DT_YEAR_MONTH_CAT</th>\n",
       "      <th>ICD9_DGNS_CD_1_CAT</th>\n",
       "      <th>ICD9_PRCDR_CD_2_CAT</th>\n",
       "      <th>ICD9_DGNS_CD_3_CAT</th>\n",
       "      <th>CLM_ADMSN_DT_CAT</th>\n",
       "      <th>NCH_BENE_IP_DDCTBL_AMT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.970000e+14</td>\n",
       "      <td>4000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>2074.0</td>\n",
       "      <td>2074.0</td>\n",
       "      <td>2758.0</td>\n",
       "      <td>826.0</td>\n",
       "      <td>1100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.960000e+14</td>\n",
       "      <td>26000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>263.0</td>\n",
       "      <td>263.0</td>\n",
       "      <td>1509.0</td>\n",
       "      <td>492.0</td>\n",
       "      <td>1068.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.970000e+14</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>1702.0</td>\n",
       "      <td>1702.0</td>\n",
       "      <td>1407.0</td>\n",
       "      <td>633.0</td>\n",
       "      <td>1068.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.960000e+14</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>653.0</td>\n",
       "      <td>653.0</td>\n",
       "      <td>1768.0</td>\n",
       "      <td>650.0</td>\n",
       "      <td>1068.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.960000e+14</td>\n",
       "      <td>16000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>873.0</td>\n",
       "      <td>873.0</td>\n",
       "      <td>783.0</td>\n",
       "      <td>932.0</td>\n",
       "      <td>1100.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         CLM_ID  CLM_PMT_AMT  CLM_PASS_THRU_PER_DIEM_AMT  CLM_UTLZTN_DAY_CNT  \\\n",
       "0  1.970000e+14       4000.0                         0.0                 1.0   \n",
       "1  1.960000e+14      26000.0                         0.0                 6.0   \n",
       "2  1.970000e+14       5000.0                         0.0                 2.0   \n",
       "3  1.960000e+14       5000.0                         0.0                 3.0   \n",
       "4  1.960000e+14      16000.0                         0.0                 5.0   \n",
       "\n",
       "   CLM_YEAR_MONTH_CLM_FROM_DT_CAT  CLM_YEAR_MONTH_CLM_THRU_DT_CAT  \\\n",
       "0                            28.0                            26.0   \n",
       "1                            17.0                            15.0   \n",
       "2                            21.0                            20.0   \n",
       "3                            22.0                            20.0   \n",
       "4                            31.0                            30.0   \n",
       "\n",
       "   NCH_BENE_DSCHRG_DT_YEAR_MONTH_CAT  ICD9_DGNS_CD_1_CAT  ICD9_PRCDR_CD_2_CAT  \\\n",
       "0                               26.0              2074.0               2074.0   \n",
       "1                               15.0               263.0                263.0   \n",
       "2                               20.0              1702.0               1702.0   \n",
       "3                               20.0               653.0                653.0   \n",
       "4                               30.0               873.0                873.0   \n",
       "\n",
       "   ICD9_DGNS_CD_3_CAT  CLM_ADMSN_DT_CAT  NCH_BENE_IP_DDCTBL_AMT  \n",
       "0              2758.0             826.0                  1100.0  \n",
       "1              1509.0             492.0                  1068.0  \n",
       "2              1407.0             633.0                  1068.0  \n",
       "3              1768.0             650.0                  1068.0  \n",
       "4               783.0             932.0                  1100.0  "
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_no_nan.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CLM_ID                               0\n",
       "CLM_PMT_AMT                          0\n",
       "CLM_PASS_THRU_PER_DIEM_AMT           0\n",
       "CLM_UTLZTN_DAY_CNT                   0\n",
       "CLM_YEAR_MONTH_CLM_FROM_DT_CAT       0\n",
       "CLM_YEAR_MONTH_CLM_THRU_DT_CAT       0\n",
       "NCH_BENE_DSCHRG_DT_YEAR_MONTH_CAT    0\n",
       "ICD9_DGNS_CD_1_CAT                   0\n",
       "ICD9_PRCDR_CD_2_CAT                  0\n",
       "ICD9_DGNS_CD_3_CAT                   0\n",
       "CLM_ADMSN_DT_CAT                     0\n",
       "NCH_BENE_IP_DDCTBL_AMT               0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_no_nan.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train1 = data_no_nan.drop(['NCH_BENE_IP_DDCTBL_AMT'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "labels = data_no_nan['NCH_BENE_IP_DDCTBL_AMT']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cross_validation import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train , x_test , y_train , y_test = train_test_split(train1 , labels , test_size = 0.10,random_state =2)\n",
    "# from sklearn.preprocessing import StandardScaler\n",
    "# sc = StandardScaler()\n",
    "# x_train = sc.fit_transform(x_train)\n",
    "# x_test = sc.transform(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg = LinearRegression()\n",
    "reg.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.93443428263056905"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg.score(x_test,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn import ensemble\n",
    "params = {'n_estimators': 100, 'max_depth': 2, 'min_samples_split': 4,\n",
    "          'learning_rate': 0.01, 'loss': 'ls'}\n",
    "clf = ensemble.GradientBoostingRegressor(**params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingRegressor(alpha=0.9, criterion='friedman_mse', init=None,\n",
       "             learning_rate=0.01, loss='ls', max_depth=2, max_features=None,\n",
       "             max_leaf_nodes=None, min_impurity_split=1e-07,\n",
       "             min_samples_leaf=1, min_samples_split=4,\n",
       "             min_weight_fraction_leaf=0.0, n_estimators=100,\n",
       "             presort='auto', random_state=None, subsample=1.0, verbose=0,\n",
       "             warm_start=False)"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.82843922745786591"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.score(x_test,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y_pred = reg.predict(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1041.20088335,  1065.76994519,  1078.59928637, ...,  1070.62606829,\n",
       "        1029.4036199 ,  1063.32633487])"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CLM_ID</th>\n",
       "      <th>CLM_PMT_AMT</th>\n",
       "      <th>CLM_PASS_THRU_PER_DIEM_AMT</th>\n",
       "      <th>CLM_UTLZTN_DAY_CNT</th>\n",
       "      <th>CLM_YEAR_MONTH_CLM_FROM_DT_CAT</th>\n",
       "      <th>CLM_YEAR_MONTH_CLM_THRU_DT_CAT</th>\n",
       "      <th>NCH_BENE_DSCHRG_DT_YEAR_MONTH_CAT</th>\n",
       "      <th>ICD9_DGNS_CD_1_CAT</th>\n",
       "      <th>ICD9_PRCDR_CD_2_CAT</th>\n",
       "      <th>ICD9_DGNS_CD_3_CAT</th>\n",
       "      <th>CLM_ADMSN_DT_CAT</th>\n",
       "      <th>NCH_BENE_IP_DDCTBL_AMT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.970000e+14</td>\n",
       "      <td>4000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>2074.0</td>\n",
       "      <td>2074.0</td>\n",
       "      <td>2758.0</td>\n",
       "      <td>826.0</td>\n",
       "      <td>1100.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         CLM_ID  CLM_PMT_AMT  CLM_PASS_THRU_PER_DIEM_AMT  CLM_UTLZTN_DAY_CNT  \\\n",
       "0  1.970000e+14       4000.0                         0.0                 1.0   \n",
       "\n",
       "   CLM_YEAR_MONTH_CLM_FROM_DT_CAT  CLM_YEAR_MONTH_CLM_THRU_DT_CAT  \\\n",
       "0                            28.0                            26.0   \n",
       "\n",
       "   NCH_BENE_DSCHRG_DT_YEAR_MONTH_CAT  ICD9_DGNS_CD_1_CAT  ICD9_PRCDR_CD_2_CAT  \\\n",
       "0                               26.0              2074.0               2074.0   \n",
       "\n",
       "   ICD9_DGNS_CD_3_CAT  CLM_ADMSN_DT_CAT  NCH_BENE_IP_DDCTBL_AMT  \n",
       "0              2758.0             826.0                  1100.0  "
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_no_nan.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "new_pred = reg.predict(np.array([[1.970000e+14,260000,3,6,17,15,15,263,263,1509,492]]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1061.71889064])"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "t_sc = np.zeros((params['n_estimators']),dtype=np.float64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y_pred = reg.predict(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "for i,y_pred in enumerate(clf.staged_predict(x_test)):\n",
    "    t_sc[i]=clf.loss_(y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "testsc = np.arange((params['n_estimators']))+1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVwAAAFyCAYAAAC5sGJqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAhCWRNZwL60XAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ba40b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 6))\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(testsc,clf.train_score_,'b-',label= 'Set dev train')\n",
    "plt.title(\"Training loss-Red vs Testing loss-Blue\")\n",
    "plt.plot(testsc,t_sc,'r-',label = 'set dev test')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1061.7188906423241"
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_pred[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1037.82249942,  1065.13608542,  1065.13608542, ...,  1065.13608542,\n",
       "        1037.82249942,  1065.13608542])"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.externals import joblib\n",
    "clf = joblib.load('SCIO-NCH_BENE_IP_DDCTBL_AMT_PREDICTOR.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction = clf.predict(np.array([[1.970000e+14,260000,3,6,17,15,15,263,263,1509,492]]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1068.0"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(prediction[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction = clf.predict(np.array([[1.970000e+14,260000,3,6,17,15,15,263,263,1509,492]]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1068.0"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(prediction[0])"
   ]
  },
  {
   "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}