71_xgb_onehot_dgi_label.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4c65f557-651f-4958-a567-a1b6b5a82b2b",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Modify the file 00_setup.py to define input/output file paths on your system\n",
    "# The information in 00_setup.py will be used across notebooks\n",
    "from importlib.machinery import SourceFileLoader\n",
    "setup = SourceFileLoader(\"setup\", \"./00_setup.py\").load_module()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b01d0fdb-d4bf-465a-8541-997addc50b0c",
   "metadata": {},
   "source": [
    "# XGBoost with one-hot NAICS Clusters (DGI Label)\n",
    "XGBoost model, one hot encoded values from k means clustering"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56161f89-1f54-47f9-860d-3f945a941c1f",
   "metadata": {},
   "source": [
    "*This script takes about 5 minutes on my MacBook Air*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ec6b1778-7abc-4129-ab5f-ac2992bd8ca6",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import xgboost as xgb\n",
    "import warnings\n",
    "import matplotlib.pyplot as plt\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "785af927-a634-4e34-84c4-7a532236f2ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "fd09527b-32dd-47fa-b62f-f9307f89f553",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sklearn as sk\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from sklearn.inspection import permutation_importance, partial_dependence\n",
    "from scikitplot.metrics import plot_lift_curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0732ba40-8318-457b-b29f-f9b6de3b70aa",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sba_gnn.sba_gnn import sg_plot "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d14a5947-6599-4253-acf3-91ddeffdedd3",
   "metadata": {},
   "source": [
    "## Import Processed Datasets"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b0e80daf-7e62-47ae-b9f4-e04468135f3e",
   "metadata": {},
   "source": [
    "##### Get loans data, append clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f6bc5cc5-b635-420f-852c-6cc9bba68a52",
   "metadata": {},
   "outputs": [],
   "source": [
    "sba_loans = pd.read_parquet(Path(setup.temp_path).joinpath('01_DATA_transformed.parquet'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "24785791-96b2-42a3-b6bb-307a60291c06",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['NAICS', 'NAICS_2', 'NAICS_sector', 'NAICS_sector_desc', 'NAICS_num',\n",
       "       'menc_NAICS', 'cenc_NAICS'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sba_loans.filter(like='NAICS').columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "5d9230bf-f527-42c9-832a-e353f32487ab",
   "metadata": {},
   "outputs": [],
   "source": [
    "embed_df = pd.read_parquet(Path(setup.temp_path).joinpath('70_DATA_embeddings_tsne_naics.parquet'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f61f3bb8-2f9b-4d2d-a85d-c9c932521b47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clus_cols = [c for c in embed_df if c.startswith('NC__')]\n",
    "len(clus_cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3533831b-fe3e-46e7-bc21-b0d6f4fb689d",
   "metadata": {},
   "outputs": [],
   "source": [
    "sba_loans = sba_loans.drop(columns = clus_cols, errors='ignore') \\\n",
    "    .merge(embed_df[clus_cols + ['NAICS_orig']].rename(columns={'NAICS_orig':'NAICS'}),\n",
    "           on='NAICS', how='left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "042aae2d-f88d-4060-994f-629349a5bff1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NC__0</th>\n",
       "      <th>NC__1</th>\n",
       "      <th>NC__2</th>\n",
       "      <th>NC__3</th>\n",
       "      <th>NC__4</th>\n",
       "      <th>NC__5</th>\n",
       "      <th>NC__6</th>\n",
       "      <th>NC__7</th>\n",
       "      <th>NC__8</th>\n",
       "      <th>NC__9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>688081.000000</td>\n",
       "      <td>688081.000000</td>\n",
       "      <td>688081.000000</td>\n",
       "      <td>688081.000000</td>\n",
       "      <td>688081.000000</td>\n",
       "      <td>688081.000000</td>\n",
       "      <td>688081.000000</td>\n",
       "      <td>688081.000000</td>\n",
       "      <td>688081.000000</td>\n",
       "      <td>688081.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.092204</td>\n",
       "      <td>0.209945</td>\n",
       "      <td>0.106208</td>\n",
       "      <td>0.064456</td>\n",
       "      <td>0.203640</td>\n",
       "      <td>0.081368</td>\n",
       "      <td>0.019261</td>\n",
       "      <td>0.074751</td>\n",
       "      <td>0.096948</td>\n",
       "      <td>0.051218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.289314</td>\n",
       "      <td>0.407269</td>\n",
       "      <td>0.308104</td>\n",
       "      <td>0.245564</td>\n",
       "      <td>0.402705</td>\n",
       "      <td>0.273400</td>\n",
       "      <td>0.137440</td>\n",
       "      <td>0.262990</td>\n",
       "      <td>0.295887</td>\n",
       "      <td>0.220442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               NC__0          NC__1          NC__2          NC__3  \\\n",
       "count  688081.000000  688081.000000  688081.000000  688081.000000   \n",
       "mean        0.092204       0.209945       0.106208       0.064456   \n",
       "std         0.289314       0.407269       0.308104       0.245564   \n",
       "min         0.000000       0.000000       0.000000       0.000000   \n",
       "25%         0.000000       0.000000       0.000000       0.000000   \n",
       "50%         0.000000       0.000000       0.000000       0.000000   \n",
       "75%         0.000000       0.000000       0.000000       0.000000   \n",
       "max         1.000000       1.000000       1.000000       1.000000   \n",
       "\n",
       "               NC__4          NC__5          NC__6          NC__7  \\\n",
       "count  688081.000000  688081.000000  688081.000000  688081.000000   \n",
       "mean        0.203640       0.081368       0.019261       0.074751   \n",
       "std         0.402705       0.273400       0.137440       0.262990   \n",
       "min         0.000000       0.000000       0.000000       0.000000   \n",
       "25%         0.000000       0.000000       0.000000       0.000000   \n",
       "50%         0.000000       0.000000       0.000000       0.000000   \n",
       "75%         0.000000       0.000000       0.000000       0.000000   \n",
       "max         1.000000       1.000000       1.000000       1.000000   \n",
       "\n",
       "               NC__8          NC__9  \n",
       "count  688081.000000  688081.000000  \n",
       "mean        0.096948       0.051218  \n",
       "std         0.295887       0.220442  \n",
       "min         0.000000       0.000000  \n",
       "25%         0.000000       0.000000  \n",
       "50%         0.000000       0.000000  \n",
       "75%         0.000000       0.000000  \n",
       "max         1.000000       1.000000  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sba_loans[clus_cols].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a5f6c362-c1ab-48d8-812b-2b970b06586b",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = sba_loans[sba_loans['dset'] == 'train'].copy()\n",
    "test_df = sba_loans[sba_loans['dset'] == 'test']\n",
    "val_df = sba_loans[sba_loans['dset'] == 'val']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7b1db53b-181c-450d-9f7e-be350bc3111f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(436120, 65)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2160b50-1e22-44f0-915b-0241ebcd45d4",
   "metadata": {},
   "source": [
    "#### Set the predictor features.  Include one-hot NAICS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c07ddd4e-6987-454a-926a-3b59abf6caa6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['NoEmp', 'CreateJob', 'LowDoc', 'DisbursementGross', 'new_business', 'urban_flag', 'franchise_flag', 'NC__0', 'NC__1', 'NC__2', 'NC__3', 'NC__4', 'NC__5', 'NC__6', 'NC__7', 'NC__8', 'NC__9']\n"
     ]
    }
   ],
   "source": [
    "predictor_features = setup.predictor_features + clus_cols\n",
    "print(predictor_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "2a4f4a9e-f7e5-4143-b1e4-b5fbb5f7280d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NoEmp</th>\n",
       "      <th>CreateJob</th>\n",
       "      <th>LowDoc</th>\n",
       "      <th>DisbursementGross</th>\n",
       "      <th>new_business</th>\n",
       "      <th>urban_flag</th>\n",
       "      <th>franchise_flag</th>\n",
       "      <th>NC__0</th>\n",
       "      <th>NC__1</th>\n",
       "      <th>NC__2</th>\n",
       "      <th>NC__3</th>\n",
       "      <th>NC__4</th>\n",
       "      <th>NC__5</th>\n",
       "      <th>NC__6</th>\n",
       "      <th>NC__7</th>\n",
       "      <th>NC__8</th>\n",
       "      <th>NC__9</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>NoEmp</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.062158</td>\n",
       "      <td>-0.013926</td>\n",
       "      <td>0.095658</td>\n",
       "      <td>-0.041427</td>\n",
       "      <td>0.001610</td>\n",
       "      <td>0.006409</td>\n",
       "      <td>0.025479</td>\n",
       "      <td>0.023862</td>\n",
       "      <td>-0.009680</td>\n",
       "      <td>-0.002758</td>\n",
       "      <td>-0.030946</td>\n",
       "      <td>-0.001089</td>\n",
       "      <td>-0.008447</td>\n",
       "      <td>0.005769</td>\n",
       "      <td>0.008647</td>\n",
       "      <td>-0.016860</td>\n",
       "      <td>-0.027403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CreateJob</th>\n",
       "      <td>0.062158</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.062740</td>\n",
       "      <td>0.134686</td>\n",
       "      <td>0.039302</td>\n",
       "      <td>0.009062</td>\n",
       "      <td>0.050517</td>\n",
       "      <td>0.062246</td>\n",
       "      <td>-0.010690</td>\n",
       "      <td>-0.003005</td>\n",
       "      <td>0.003270</td>\n",
       "      <td>-0.025301</td>\n",
       "      <td>-0.013564</td>\n",
       "      <td>-0.016220</td>\n",
       "      <td>-0.009872</td>\n",
       "      <td>0.035795</td>\n",
       "      <td>-0.028751</td>\n",
       "      <td>-0.010002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LowDoc</th>\n",
       "      <td>-0.013926</td>\n",
       "      <td>-0.062740</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.133141</td>\n",
       "      <td>0.153034</td>\n",
       "      <td>-0.120001</td>\n",
       "      <td>0.040841</td>\n",
       "      <td>-0.025616</td>\n",
       "      <td>0.075187</td>\n",
       "      <td>-0.045471</td>\n",
       "      <td>-0.048888</td>\n",
       "      <td>-0.063928</td>\n",
       "      <td>0.025281</td>\n",
       "      <td>0.023207</td>\n",
       "      <td>0.026131</td>\n",
       "      <td>-0.031881</td>\n",
       "      <td>0.098629</td>\n",
       "      <td>-0.079523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DisbursementGross</th>\n",
       "      <td>0.095658</td>\n",
       "      <td>0.134686</td>\n",
       "      <td>-0.133141</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.065443</td>\n",
       "      <td>-0.007016</td>\n",
       "      <td>0.085712</td>\n",
       "      <td>0.168745</td>\n",
       "      <td>0.082991</td>\n",
       "      <td>-0.043322</td>\n",
       "      <td>0.000937</td>\n",
       "      <td>-0.153383</td>\n",
       "      <td>-0.009432</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>-0.004506</td>\n",
       "      <td>0.033699</td>\n",
       "      <td>-0.065631</td>\n",
       "      <td>-0.120788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>new_business</th>\n",
       "      <td>-0.041427</td>\n",
       "      <td>0.039302</td>\n",
       "      <td>0.153034</td>\n",
       "      <td>-0.065443</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.047592</td>\n",
       "      <td>0.137325</td>\n",
       "      <td>0.018862</td>\n",
       "      <td>-0.039241</td>\n",
       "      <td>-0.003453</td>\n",
       "      <td>-0.015562</td>\n",
       "      <td>0.000489</td>\n",
       "      <td>-0.004373</td>\n",
       "      <td>0.024232</td>\n",
       "      <td>-0.025860</td>\n",
       "      <td>0.027801</td>\n",
       "      <td>0.051234</td>\n",
       "      <td>0.013878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>urban_flag</th>\n",
       "      <td>0.001610</td>\n",
       "      <td>0.009062</td>\n",
       "      <td>-0.120001</td>\n",
       "      <td>-0.007016</td>\n",
       "      <td>-0.047592</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.006588</td>\n",
       "      <td>-0.039184</td>\n",
       "      <td>-0.017330</td>\n",
       "      <td>0.006424</td>\n",
       "      <td>0.003897</td>\n",
       "      <td>0.033506</td>\n",
       "      <td>0.021874</td>\n",
       "      <td>-0.062527</td>\n",
       "      <td>0.012482</td>\n",
       "      <td>-0.007103</td>\n",
       "      <td>0.003005</td>\n",
       "      <td>0.053785</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>franchise_flag</th>\n",
       "      <td>0.006409</td>\n",
       "      <td>0.050517</td>\n",
       "      <td>0.040841</td>\n",
       "      <td>0.085712</td>\n",
       "      <td>0.137325</td>\n",
       "      <td>-0.006588</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.083091</td>\n",
       "      <td>-0.043945</td>\n",
       "      <td>-0.009953</td>\n",
       "      <td>-0.046530</td>\n",
       "      <td>-0.032679</td>\n",
       "      <td>-0.022046</td>\n",
       "      <td>-0.018551</td>\n",
       "      <td>-0.027687</td>\n",
       "      <td>0.137447</td>\n",
       "      <td>-0.021775</td>\n",
       "      <td>-0.012731</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__0</th>\n",
       "      <td>0.025479</td>\n",
       "      <td>0.062246</td>\n",
       "      <td>-0.025616</td>\n",
       "      <td>0.168745</td>\n",
       "      <td>0.018862</td>\n",
       "      <td>-0.039184</td>\n",
       "      <td>0.083091</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.171847</td>\n",
       "      <td>-0.115914</td>\n",
       "      <td>-0.087443</td>\n",
       "      <td>-0.171230</td>\n",
       "      <td>-0.091653</td>\n",
       "      <td>-0.037873</td>\n",
       "      <td>-0.091607</td>\n",
       "      <td>-0.109120</td>\n",
       "      <td>-0.081442</td>\n",
       "      <td>-0.014646</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__1</th>\n",
       "      <td>0.023862</td>\n",
       "      <td>-0.010690</td>\n",
       "      <td>0.075187</td>\n",
       "      <td>0.082991</td>\n",
       "      <td>-0.039241</td>\n",
       "      <td>-0.017330</td>\n",
       "      <td>-0.043945</td>\n",
       "      <td>-0.171847</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.179958</td>\n",
       "      <td>-0.135757</td>\n",
       "      <td>-0.265838</td>\n",
       "      <td>-0.142294</td>\n",
       "      <td>-0.058798</td>\n",
       "      <td>-0.142221</td>\n",
       "      <td>-0.169411</td>\n",
       "      <td>-0.126440</td>\n",
       "      <td>-0.123365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__2</th>\n",
       "      <td>-0.009680</td>\n",
       "      <td>-0.003005</td>\n",
       "      <td>-0.045471</td>\n",
       "      <td>-0.043322</td>\n",
       "      <td>-0.003453</td>\n",
       "      <td>0.006424</td>\n",
       "      <td>-0.009953</td>\n",
       "      <td>-0.115914</td>\n",
       "      <td>-0.179958</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.091570</td>\n",
       "      <td>-0.179312</td>\n",
       "      <td>-0.095979</td>\n",
       "      <td>-0.039660</td>\n",
       "      <td>-0.095931</td>\n",
       "      <td>-0.114270</td>\n",
       "      <td>-0.085286</td>\n",
       "      <td>0.049658</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__3</th>\n",
       "      <td>-0.002758</td>\n",
       "      <td>0.003270</td>\n",
       "      <td>-0.048888</td>\n",
       "      <td>0.000937</td>\n",
       "      <td>-0.015562</td>\n",
       "      <td>0.003897</td>\n",
       "      <td>-0.046530</td>\n",
       "      <td>-0.087443</td>\n",
       "      <td>-0.135757</td>\n",
       "      <td>-0.091570</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.135269</td>\n",
       "      <td>-0.072405</td>\n",
       "      <td>-0.029919</td>\n",
       "      <td>-0.072368</td>\n",
       "      <td>-0.086203</td>\n",
       "      <td>-0.064338</td>\n",
       "      <td>0.018097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__4</th>\n",
       "      <td>-0.030946</td>\n",
       "      <td>-0.025301</td>\n",
       "      <td>-0.063928</td>\n",
       "      <td>-0.153383</td>\n",
       "      <td>0.000489</td>\n",
       "      <td>0.033506</td>\n",
       "      <td>-0.032679</td>\n",
       "      <td>-0.171230</td>\n",
       "      <td>-0.265838</td>\n",
       "      <td>-0.179312</td>\n",
       "      <td>-0.135269</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.141782</td>\n",
       "      <td>-0.058586</td>\n",
       "      <td>-0.141710</td>\n",
       "      <td>-0.168802</td>\n",
       "      <td>-0.125986</td>\n",
       "      <td>0.129958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__5</th>\n",
       "      <td>-0.001089</td>\n",
       "      <td>-0.013564</td>\n",
       "      <td>0.025281</td>\n",
       "      <td>-0.009432</td>\n",
       "      <td>-0.004373</td>\n",
       "      <td>0.021874</td>\n",
       "      <td>-0.022046</td>\n",
       "      <td>-0.091653</td>\n",
       "      <td>-0.142294</td>\n",
       "      <td>-0.095979</td>\n",
       "      <td>-0.072405</td>\n",
       "      <td>-0.141782</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.031359</td>\n",
       "      <td>-0.075853</td>\n",
       "      <td>-0.090354</td>\n",
       "      <td>-0.067436</td>\n",
       "      <td>-0.026586</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__6</th>\n",
       "      <td>-0.008447</td>\n",
       "      <td>-0.016220</td>\n",
       "      <td>0.023207</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.024232</td>\n",
       "      <td>-0.062527</td>\n",
       "      <td>-0.018551</td>\n",
       "      <td>-0.037873</td>\n",
       "      <td>-0.058798</td>\n",
       "      <td>-0.039660</td>\n",
       "      <td>-0.029919</td>\n",
       "      <td>-0.058586</td>\n",
       "      <td>-0.031359</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.031343</td>\n",
       "      <td>-0.037335</td>\n",
       "      <td>-0.027865</td>\n",
       "      <td>-0.014604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__7</th>\n",
       "      <td>0.005769</td>\n",
       "      <td>-0.009872</td>\n",
       "      <td>0.026131</td>\n",
       "      <td>-0.004506</td>\n",
       "      <td>-0.025860</td>\n",
       "      <td>0.012482</td>\n",
       "      <td>-0.027687</td>\n",
       "      <td>-0.091607</td>\n",
       "      <td>-0.142221</td>\n",
       "      <td>-0.095931</td>\n",
       "      <td>-0.072368</td>\n",
       "      <td>-0.141710</td>\n",
       "      <td>-0.075853</td>\n",
       "      <td>-0.031343</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.090308</td>\n",
       "      <td>-0.067402</td>\n",
       "      <td>-0.047362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__8</th>\n",
       "      <td>0.008647</td>\n",
       "      <td>0.035795</td>\n",
       "      <td>-0.031881</td>\n",
       "      <td>0.033699</td>\n",
       "      <td>0.027801</td>\n",
       "      <td>-0.007103</td>\n",
       "      <td>0.137447</td>\n",
       "      <td>-0.109120</td>\n",
       "      <td>-0.169411</td>\n",
       "      <td>-0.114270</td>\n",
       "      <td>-0.086203</td>\n",
       "      <td>-0.168802</td>\n",
       "      <td>-0.090354</td>\n",
       "      <td>-0.037335</td>\n",
       "      <td>-0.090308</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.080287</td>\n",
       "      <td>0.015704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__9</th>\n",
       "      <td>-0.016860</td>\n",
       "      <td>-0.028751</td>\n",
       "      <td>0.098629</td>\n",
       "      <td>-0.065631</td>\n",
       "      <td>0.051234</td>\n",
       "      <td>0.003005</td>\n",
       "      <td>-0.021775</td>\n",
       "      <td>-0.081442</td>\n",
       "      <td>-0.126440</td>\n",
       "      <td>-0.085286</td>\n",
       "      <td>-0.064338</td>\n",
       "      <td>-0.125986</td>\n",
       "      <td>-0.067436</td>\n",
       "      <td>-0.027865</td>\n",
       "      <td>-0.067402</td>\n",
       "      <td>-0.080287</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.009301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <td>-0.027403</td>\n",
       "      <td>-0.010002</td>\n",
       "      <td>-0.079523</td>\n",
       "      <td>-0.120788</td>\n",
       "      <td>0.013878</td>\n",
       "      <td>0.053785</td>\n",
       "      <td>-0.012731</td>\n",
       "      <td>-0.014646</td>\n",
       "      <td>-0.123365</td>\n",
       "      <td>0.049658</td>\n",
       "      <td>0.018097</td>\n",
       "      <td>0.129958</td>\n",
       "      <td>-0.026586</td>\n",
       "      <td>-0.014604</td>\n",
       "      <td>-0.047362</td>\n",
       "      <td>0.015704</td>\n",
       "      <td>-0.009301</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      NoEmp  CreateJob    LowDoc  DisbursementGross  \\\n",
       "NoEmp              1.000000   0.062158 -0.013926           0.095658   \n",
       "CreateJob          0.062158   1.000000 -0.062740           0.134686   \n",
       "LowDoc            -0.013926  -0.062740  1.000000          -0.133141   \n",
       "DisbursementGross  0.095658   0.134686 -0.133141           1.000000   \n",
       "new_business      -0.041427   0.039302  0.153034          -0.065443   \n",
       "urban_flag         0.001610   0.009062 -0.120001          -0.007016   \n",
       "franchise_flag     0.006409   0.050517  0.040841           0.085712   \n",
       "NC__0              0.025479   0.062246 -0.025616           0.168745   \n",
       "NC__1              0.023862  -0.010690  0.075187           0.082991   \n",
       "NC__2             -0.009680  -0.003005 -0.045471          -0.043322   \n",
       "NC__3             -0.002758   0.003270 -0.048888           0.000937   \n",
       "NC__4             -0.030946  -0.025301 -0.063928          -0.153383   \n",
       "NC__5             -0.001089  -0.013564  0.025281          -0.009432   \n",
       "NC__6             -0.008447  -0.016220  0.023207           0.000004   \n",
       "NC__7              0.005769  -0.009872  0.026131          -0.004506   \n",
       "NC__8              0.008647   0.035795 -0.031881           0.033699   \n",
       "NC__9             -0.016860  -0.028751  0.098629          -0.065631   \n",
       "target            -0.027403  -0.010002 -0.079523          -0.120788   \n",
       "\n",
       "                   new_business  urban_flag  franchise_flag     NC__0  \\\n",
       "NoEmp                 -0.041427    0.001610        0.006409  0.025479   \n",
       "CreateJob              0.039302    0.009062        0.050517  0.062246   \n",
       "LowDoc                 0.153034   -0.120001        0.040841 -0.025616   \n",
       "DisbursementGross     -0.065443   -0.007016        0.085712  0.168745   \n",
       "new_business           1.000000   -0.047592        0.137325  0.018862   \n",
       "urban_flag            -0.047592    1.000000       -0.006588 -0.039184   \n",
       "franchise_flag         0.137325   -0.006588        1.000000  0.083091   \n",
       "NC__0                  0.018862   -0.039184        0.083091  1.000000   \n",
       "NC__1                 -0.039241   -0.017330       -0.043945 -0.171847   \n",
       "NC__2                 -0.003453    0.006424       -0.009953 -0.115914   \n",
       "NC__3                 -0.015562    0.003897       -0.046530 -0.087443   \n",
       "NC__4                  0.000489    0.033506       -0.032679 -0.171230   \n",
       "NC__5                 -0.004373    0.021874       -0.022046 -0.091653   \n",
       "NC__6                  0.024232   -0.062527       -0.018551 -0.037873   \n",
       "NC__7                 -0.025860    0.012482       -0.027687 -0.091607   \n",
       "NC__8                  0.027801   -0.007103        0.137447 -0.109120   \n",
       "NC__9                  0.051234    0.003005       -0.021775 -0.081442   \n",
       "target                 0.013878    0.053785       -0.012731 -0.014646   \n",
       "\n",
       "                      NC__1     NC__2     NC__3     NC__4     NC__5     NC__6  \\\n",
       "NoEmp              0.023862 -0.009680 -0.002758 -0.030946 -0.001089 -0.008447   \n",
       "CreateJob         -0.010690 -0.003005  0.003270 -0.025301 -0.013564 -0.016220   \n",
       "LowDoc             0.075187 -0.045471 -0.048888 -0.063928  0.025281  0.023207   \n",
       "DisbursementGross  0.082991 -0.043322  0.000937 -0.153383 -0.009432  0.000004   \n",
       "new_business      -0.039241 -0.003453 -0.015562  0.000489 -0.004373  0.024232   \n",
       "urban_flag        -0.017330  0.006424  0.003897  0.033506  0.021874 -0.062527   \n",
       "franchise_flag    -0.043945 -0.009953 -0.046530 -0.032679 -0.022046 -0.018551   \n",
       "NC__0             -0.171847 -0.115914 -0.087443 -0.171230 -0.091653 -0.037873   \n",
       "NC__1              1.000000 -0.179958 -0.135757 -0.265838 -0.142294 -0.058798   \n",
       "NC__2             -0.179958  1.000000 -0.091570 -0.179312 -0.095979 -0.039660   \n",
       "NC__3             -0.135757 -0.091570  1.000000 -0.135269 -0.072405 -0.029919   \n",
       "NC__4             -0.265838 -0.179312 -0.135269  1.000000 -0.141782 -0.058586   \n",
       "NC__5             -0.142294 -0.095979 -0.072405 -0.141782  1.000000 -0.031359   \n",
       "NC__6             -0.058798 -0.039660 -0.029919 -0.058586 -0.031359  1.000000   \n",
       "NC__7             -0.142221 -0.095931 -0.072368 -0.141710 -0.075853 -0.031343   \n",
       "NC__8             -0.169411 -0.114270 -0.086203 -0.168802 -0.090354 -0.037335   \n",
       "NC__9             -0.126440 -0.085286 -0.064338 -0.125986 -0.067436 -0.027865   \n",
       "target            -0.123365  0.049658  0.018097  0.129958 -0.026586 -0.014604   \n",
       "\n",
       "                      NC__7     NC__8     NC__9    target  \n",
       "NoEmp              0.005769  0.008647 -0.016860 -0.027403  \n",
       "CreateJob         -0.009872  0.035795 -0.028751 -0.010002  \n",
       "LowDoc             0.026131 -0.031881  0.098629 -0.079523  \n",
       "DisbursementGross -0.004506  0.033699 -0.065631 -0.120788  \n",
       "new_business      -0.025860  0.027801  0.051234  0.013878  \n",
       "urban_flag         0.012482 -0.007103  0.003005  0.053785  \n",
       "franchise_flag    -0.027687  0.137447 -0.021775 -0.012731  \n",
       "NC__0             -0.091607 -0.109120 -0.081442 -0.014646  \n",
       "NC__1             -0.142221 -0.169411 -0.126440 -0.123365  \n",
       "NC__2             -0.095931 -0.114270 -0.085286  0.049658  \n",
       "NC__3             -0.072368 -0.086203 -0.064338  0.018097  \n",
       "NC__4             -0.141710 -0.168802 -0.125986  0.129958  \n",
       "NC__5             -0.075853 -0.090354 -0.067436 -0.026586  \n",
       "NC__6             -0.031343 -0.037335 -0.027865 -0.014604  \n",
       "NC__7              1.000000 -0.090308 -0.067402 -0.047362  \n",
       "NC__8             -0.090308  1.000000 -0.080287  0.015704  \n",
       "NC__9             -0.067402 -0.080287  1.000000 -0.009301  \n",
       "target            -0.047362  0.015704 -0.009301  1.000000  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df[predictor_features + ['target']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "1f8c475d-7ead-4f92-a19b-326343528f59",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(Path(setup.temp_path).joinpath('71_DATA_features_predict.pkl'), 'wb') as fout:\n",
    "    pickle.dump(predictor_features, fout)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b204a92-d109-4f6d-87a5-3d5dc05d2e5f",
   "metadata": {},
   "source": [
    "## Model Fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "c07828e0-5f5e-433f-9d96-fe78e9d668ff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'subsample': 0.8,\n",
       " 'scale_pos_weight': 1.0,\n",
       " 'reg_lambda': 0.01,\n",
       " 'reg_alpha': 0.001,\n",
       " 'min_child_weight': 50,\n",
       " 'max_depth': 8,\n",
       " 'learning_rate': 0.1,\n",
       " 'gamma': 2}"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_params_df = pd.read_csv(Path(setup.temp_path).joinpath('03_REPORT_fit_parameter_selected.csv'))\n",
    "best_params = best_params_df.to_dict(orient='records')[0]\n",
    "best_params"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09bb1219-1d1e-432e-a2ca-be94168cf7ba",
   "metadata": {},
   "source": [
    "##### Fit using pre-determined parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "3eeec4bc-9ca2-4d11-87b8-a81a7fa93383",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_params = dict(best_params, \n",
    "                   **{'objective':\"binary:logistic\", 'n_estimators':setup.xgb_n_estimators})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "7cb431ab-94aa-4922-8107-e97f0e50314f",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_model = xgb.XGBClassifier(objective=\"binary:logistic\",\n",
    "                             n_estimators = setup.xgb_n_estimators)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "1e17a637-89fd-4ee7-8e4f-a219133d2d75",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_model = xgb_model.set_params(**xgb_params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "c3d7133b-075a-48a2-b9c8-c004dab7be61",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0]\tvalidation_0-logloss:0.50198\tvalidation_1-logloss:0.50543\n",
      "[1]\tvalidation_0-logloss:0.49565\tvalidation_1-logloss:0.49917\n",
      "[2]\tvalidation_0-logloss:0.49045\tvalidation_1-logloss:0.49405\n",
      "[3]\tvalidation_0-logloss:0.48593\tvalidation_1-logloss:0.48960\n",
      "[4]\tvalidation_0-logloss:0.48216\tvalidation_1-logloss:0.48598\n",
      "[5]\tvalidation_0-logloss:0.47893\tvalidation_1-logloss:0.48283\n",
      "[6]\tvalidation_0-logloss:0.47605\tvalidation_1-logloss:0.48004\n",
      "[7]\tvalidation_0-logloss:0.47357\tvalidation_1-logloss:0.47764\n",
      "[8]\tvalidation_0-logloss:0.47149\tvalidation_1-logloss:0.47564\n",
      "[9]\tvalidation_0-logloss:0.46972\tvalidation_1-logloss:0.47395\n",
      "[10]\tvalidation_0-logloss:0.46811\tvalidation_1-logloss:0.47241\n",
      "[11]\tvalidation_0-logloss:0.46681\tvalidation_1-logloss:0.47118\n",
      "[12]\tvalidation_0-logloss:0.46566\tvalidation_1-logloss:0.47013\n",
      "[13]\tvalidation_0-logloss:0.46470\tvalidation_1-logloss:0.46924\n",
      "[14]\tvalidation_0-logloss:0.46371\tvalidation_1-logloss:0.46832\n",
      "[15]\tvalidation_0-logloss:0.46291\tvalidation_1-logloss:0.46761\n",
      "[16]\tvalidation_0-logloss:0.46222\tvalidation_1-logloss:0.46699\n",
      "[17]\tvalidation_0-logloss:0.46161\tvalidation_1-logloss:0.46648\n",
      "[18]\tvalidation_0-logloss:0.46098\tvalidation_1-logloss:0.46591\n",
      "[19]\tvalidation_0-logloss:0.46045\tvalidation_1-logloss:0.46544\n",
      "[20]\tvalidation_0-logloss:0.46006\tvalidation_1-logloss:0.46510\n",
      "[21]\tvalidation_0-logloss:0.45968\tvalidation_1-logloss:0.46478\n",
      "[22]\tvalidation_0-logloss:0.45936\tvalidation_1-logloss:0.46452\n",
      "[23]\tvalidation_0-logloss:0.45907\tvalidation_1-logloss:0.46428\n",
      "[24]\tvalidation_0-logloss:0.45880\tvalidation_1-logloss:0.46406\n",
      "[25]\tvalidation_0-logloss:0.45851\tvalidation_1-logloss:0.46378\n",
      "[26]\tvalidation_0-logloss:0.45830\tvalidation_1-logloss:0.46362\n",
      "[27]\tvalidation_0-logloss:0.45808\tvalidation_1-logloss:0.46344\n",
      "[28]\tvalidation_0-logloss:0.45786\tvalidation_1-logloss:0.46327\n",
      "[29]\tvalidation_0-logloss:0.45767\tvalidation_1-logloss:0.46312\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=2, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=0.1, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=8, max_leaves=None,\n",
       "              min_child_weight=50, missing=nan, monotone_constraints=None,\n",
       "              multi_strategy=None, n_estimators=30, n_jobs=None,\n",
       "              num_parallel_tree=None, random_state=None, ...)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=2, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=0.1, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=8, max_leaves=None,\n",
       "              min_child_weight=50, missing=nan, monotone_constraints=None,\n",
       "              multi_strategy=None, n_estimators=30, n_jobs=None,\n",
       "              num_parallel_tree=None, random_state=None, ...)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=2, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=0.1, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=8, max_leaves=None,\n",
       "              min_child_weight=50, missing=nan, monotone_constraints=None,\n",
       "              multi_strategy=None, n_estimators=30, n_jobs=None,\n",
       "              num_parallel_tree=None, random_state=None, ...)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb_model.fit(train_df[predictor_features], train_df['target'],\n",
    "              eval_set = [(train_df[predictor_features], train_df['target']),\n",
    "                         (val_df[predictor_features], val_df['target'])])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0913a102-a730-4b94-9977-e584df764d64",
   "metadata": {},
   "source": [
    "##### Save the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "af21dd7c-09d4-407f-9960-0748ba8f2247",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_model.save_model(Path(setup.temp_path).joinpath('71_MODEL_xgboost.json'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "8e5d3e7d-0bd5-4252-b287-c9cc65f7a132",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(Path(setup.temp_path).joinpath('71_MODEL_xgboost.pkl'), 'wb') as po:\n",
    "    pickle.dump(xgb_model, po)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f34ad059-70dc-4672-adff-0d961fc74b07",
   "metadata": {},
   "source": [
    "## Probability Predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "6a7c8509-0109-4f8e-a466-d6e21164012c",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_df = pd.concat([sba_loans[['LoanNr_ChkDgt', 'dset', 'target', 'dset_naics_holdout']].reset_index(),\n",
    "                     pd.DataFrame({'predict_prob':\n",
    "                                   xgb_model.predict_proba(sba_loans[predictor_features])[:,1]})],\n",
    "                    axis=1) \\\n",
    "    .set_index('index')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "7084012d-b68a-4ca4-b714-160f8d564500",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>max</th>\n",
       "      <th>median</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.187850</td>\n",
       "      <td>0.107819</td>\n",
       "      <td>0.01568</td>\n",
       "      <td>0.675339</td>\n",
       "      <td>0.184571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.271484</td>\n",
       "      <td>0.095430</td>\n",
       "      <td>0.01568</td>\n",
       "      <td>0.697918</td>\n",
       "      <td>0.283431</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            mean       std      min       max    median\n",
       "target                                                 \n",
       "0       0.187850  0.107819  0.01568  0.675339  0.184571\n",
       "1       0.271484  0.095430  0.01568  0.697918  0.283431"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_df.groupby('target')['predict_prob'].agg(['mean', 'std', 'min', 'max', 'median'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d33cb4dc-6c89-4a37-98c7-d3a666abc4a5",
   "metadata": {},
   "source": [
    "## Decision Threshold\n",
    "Max f1, on training data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "d7727498-825f-4f53-882f-cfcce730b1fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_train_df = pred_df[pred_df['dset'] == 'train']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "7b553798-eead-48f8-b14b-5f036d9183d9",
   "metadata": {},
   "outputs": [],
   "source": [
    "thresh_tune_data = sg_plot.get_f1_frame(pred_train_df['target'], \n",
    "                                        pred_train_df['predict_prob'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "7a1cde01-a30b-475f-8fd7-409b13946c6d",
   "metadata": {},
   "outputs": [],
   "source": [
    "thresh_tune_data.to_csv(Path(setup.temp_path).joinpath('71_REPORT_thresh_tune.csv'), index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "947a487c-e71d-433c-bbf0-2ed7e57a908a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.22\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>thresh</th>\n",
       "      <th>f1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.22</td>\n",
       "      <td>0.437708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.24</td>\n",
       "      <td>0.437467</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    thresh        f1\n",
       "11    0.22  0.437708\n",
       "12    0.24  0.437467"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "thresh_head = thresh_tune_data.sort_values('f1', ascending=False).head(2)\n",
    "best_thresh = thresh_head['thresh'].iloc[0]\n",
    "print(best_thresh)\n",
    "thresh_head"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1ee5330e-c1a7-4310-afbe-be9bc0d26f43",
   "metadata": {},
   "source": [
    "##### Apply decision threshold\n",
    "All datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "f0d4c1cf-6d5b-49e9-9123-a067697a9e39",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_df['predict_bin'] = np.where(pred_df['predict_prob'] >= best_thresh, 1, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "8253c32c-0438-47b9-8570-7cde6f7f5590",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "predict_bin\n",
       "0    356422\n",
       "1    331659\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_df['predict_bin'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "dbb46f4a-a5ef-4fd6-9da4-30de4a00334d",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_df.to_parquet(Path(setup.temp_path).joinpath('71_DATA_combined_predictions.parquet'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ab860d0-eddc-4f63-b853-88da15d451f4",
   "metadata": {},
   "source": [
    "## Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "f2801964-124e-425f-b4cb-9bcf1cf3f04a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>dset</th>\n",
       "      <th>accuracy_score</th>\n",
       "      <th>f1_score</th>\n",
       "      <th>precision_score</th>\n",
       "      <th>recall_score</th>\n",
       "      <th>average_precision_score</th>\n",
       "      <th>roc_auc_score</th>\n",
       "      <th>dset_naics_holdout</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>test</td>\n",
       "      <td>0.623726</td>\n",
       "      <td>0.433373</td>\n",
       "      <td>0.307327</td>\n",
       "      <td>0.734699</td>\n",
       "      <td>0.358936</td>\n",
       "      <td>0.722671</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>train</td>\n",
       "      <td>0.611536</td>\n",
       "      <td>0.437708</td>\n",
       "      <td>0.310975</td>\n",
       "      <td>0.738790</td>\n",
       "      <td>0.363729</td>\n",
       "      <td>0.715256</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>val</td>\n",
       "      <td>0.610140</td>\n",
       "      <td>0.437677</td>\n",
       "      <td>0.312299</td>\n",
       "      <td>0.731253</td>\n",
       "      <td>0.362558</td>\n",
       "      <td>0.710277</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>test</td>\n",
       "      <td>0.609759</td>\n",
       "      <td>0.436635</td>\n",
       "      <td>0.310527</td>\n",
       "      <td>0.735213</td>\n",
       "      <td>0.359239</td>\n",
       "      <td>0.711826</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>test</td>\n",
       "      <td>0.643793</td>\n",
       "      <td>0.428162</td>\n",
       "      <td>0.302254</td>\n",
       "      <td>0.733863</td>\n",
       "      <td>0.358599</td>\n",
       "      <td>0.737109</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    dset  accuracy_score  f1_score  precision_score  recall_score  \\\n",
       "0   test        0.623726  0.433373         0.307327      0.734699   \n",
       "1  train        0.611536  0.437708         0.310975      0.738790   \n",
       "2    val        0.610140  0.437677         0.312299      0.731253   \n",
       "0   test        0.609759  0.436635         0.310527      0.735213   \n",
       "1   test        0.643793  0.428162         0.302254      0.733863   \n",
       "\n",
       "   average_precision_score  roc_auc_score  dset_naics_holdout  \n",
       "0                 0.358936       0.722671                 NaN  \n",
       "1                 0.363729       0.715256                 NaN  \n",
       "2                 0.362558       0.710277                 NaN  \n",
       "0                 0.359239       0.711826                 0.0  \n",
       "1                 0.358599       0.737109                 1.0  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics_dset_df = pred_df.groupby('dset') \\\n",
    "    .apply(lambda x: sg_plot.dset_metrics(x.target, x.predict_bin, x.predict_prob)) \\\n",
    "    .reset_index()\n",
    "metrics_test_df = pred_df[pred_df['dset'] == 'test'] \\\n",
    "    .groupby(['dset', 'dset_naics_holdout']) \\\n",
    "    .apply(lambda x: sg_plot.dset_metrics(x.target, x.predict_bin, x.predict_prob)) \\\n",
    "    .reset_index()\n",
    "metrics_df = pd.concat([metrics_dset_df, metrics_test_df])\n",
    "metrics_df.to_csv(Path(setup.temp_path).joinpath('71_REPORT_metrics.csv'), index=True)\n",
    "metrics_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3d906270-4264-4b0b-9f50-dcef8f75a5fd",
   "metadata": {},
   "source": [
    "## Lift\n",
    "Test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "82d9c0ab-8989-450d-a2b1-7997ca636052",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "pred_test_df = pred_df[pred_df['dset'] == 'test']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "c01f5279-1151-4a8d-9b21-e12d4092f3c1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_lift_curve(pred_test_df['target'], \n",
    "                pd.concat([1-pred_test_df['predict_prob'], pred_test_df['predict_prob']], axis=1), \n",
    "                title='Lift Curve', ax=None, figsize=(4,3), \n",
    "                title_fontsize='large', text_fontsize='medium')\n",
    "plt.gca().legend(loc='upper right')\n",
    "plt.gcf().savefig(Path(setup.temp_path).joinpath('71_PLOT_lift_curve.png'),bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "04994f41-3c98-412a-bb2f-11e5ae22b432",
   "metadata": {},
   "source": [
    "## Feature Importance "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "926af7df-e407-484b-bd86-ae7d76a46392",
   "metadata": {},
   "source": [
    "##### XGBoost gain"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "ec3f6e1f-2b29-4b46-ac12-7a8451a3a45b",
   "metadata": {},
   "outputs": [],
   "source": [
    "feat_impgain = pd.DataFrame(xgb_model.get_booster().get_score(importance_type='gain'),\n",
    "                                     index=['importance']).T \\\n",
    "    .sort_values('importance', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "9586c637-7add-4401-b61c-37abd55b11b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "feat_impgain.to_csv(Path(setup.temp_path).joinpath('71_REPORT_importance_gain.csv'), index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "311e111f-d8e5-4a52-8adc-6a7c5f9daa8f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>urban_flag</th>\n",
       "      <td>652.829590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LowDoc</th>\n",
       "      <td>138.415176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DisbursementGross</th>\n",
       "      <td>70.928558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC__4</th>\n",
       "      <td>64.206215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>new_business</th>\n",
       "      <td>28.364401</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   importance\n",
       "urban_flag         652.829590\n",
       "LowDoc             138.415176\n",
       "DisbursementGross   70.928558\n",
       "NC__4               64.206215\n",
       "new_business        28.364401"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feat_impgain.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "9c88564a-7289-4c49-b2cd-b97338d1b042",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "feat_impgain.plot(kind='barh', legend=None)\n",
    "plt.gca().invert_yaxis()\n",
    "plt.xlabel('gain importance')\n",
    "plt.gcf().savefig(Path(setup.temp_path).joinpath('71_PLOT_importance_gain.png'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27e5fb90-f305-40ff-acaf-285d7d056c15",
   "metadata": {},
   "source": [
    "##### Permutation Importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "e2c1c123-20d0-423d-b684-585c0a5d6d2b",
   "metadata": {},
   "outputs": [],
   "source": [
    "perm_importance = permutation_importance(xgb_model, \n",
    "                                         train_df[predictor_features],  train_df['target'],\n",
    "                                         n_repeats=30, scoring = 'average_precision')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "ab0df42e-af4e-46b1-b0cf-630877137678",
   "metadata": {},
   "outputs": [],
   "source": [
    "perm_importance_df = pd.DataFrame({'features':predictor_features,\n",
    "                                   'importance': perm_importance['importances_mean']}) \n",
    "perm_importance_df['abs_importance'] = np.abs(perm_importance_df['importance'])\n",
    "perm_importance_df = perm_importance_df.sort_values('abs_importance', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "0cbc5d84-6775-404f-a685-44f5be5cd481",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>features</th>\n",
       "      <th>importance</th>\n",
       "      <th>abs_importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DisbursementGross</td>\n",
       "      <td>0.080222</td>\n",
       "      <td>0.080222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>urban_flag</td>\n",
       "      <td>0.038682</td>\n",
       "      <td>0.038682</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CreateJob</td>\n",
       "      <td>0.027264</td>\n",
       "      <td>0.027264</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LowDoc</td>\n",
       "      <td>0.016187</td>\n",
       "      <td>0.016187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>NC__4</td>\n",
       "      <td>0.013909</td>\n",
       "      <td>0.013909</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             features  importance  abs_importance\n",
       "3   DisbursementGross    0.080222        0.080222\n",
       "5          urban_flag    0.038682        0.038682\n",
       "1           CreateJob    0.027264        0.027264\n",
       "2              LowDoc    0.016187        0.016187\n",
       "11              NC__4    0.013909        0.013909"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "perm_importance_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "38059608-2f72-491f-86b3-0c1418b83531",
   "metadata": {},
   "outputs": [],
   "source": [
    "perm_importance_df \\\n",
    "    .to_csv(Path(setup.temp_path).joinpath('71_DATA_importance_perturb.csv'),\n",
    "            index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "c27499c8-065b-4143-a7dc-bce301bf3edf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "perm_importance_df.head(10).plot(kind='barh', x='features', y='importance', legend=None)\n",
    "plt.gca().invert_yaxis()\n",
    "plt.xlabel('permutation importance')\n",
    "plt.ylabel(None)\n",
    "plt.gcf().savefig(Path(setup.temp_path).joinpath('71_PLOT_importance_perturb.png'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6fd44a7c-bcf5-4284-8b9f-b575fa523ff3",
   "metadata": {},
   "source": [
    "## PDP Plot - One Hot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "0f5a4774-bf8b-4c7f-9003-e1bcb4c101f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "naics_features = list(train_df.filter(like='NC_').columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "730411c5-81d3-4911-ad8f-8743dc1617e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "85ab4bf4-0f82-4121-842d-fd22164982cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x900 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(int(np.ceil(len(naics_features)/2)), 2, sharex=True, sharey=True, figsize=(6, 9))\n",
    "for i in range(len(naics_features)):\n",
    "    col = int(i % 2)\n",
    "    row = int(np.floor(i / 2))\n",
    "    results = partial_dependence(xgb_model, train_df[predictor_features], \n",
    "                                 [naics_features[i]])\n",
    "    pdp_df = pd.DataFrame({'grid_values':results[\"grid_values\"][0],\n",
    "                       'pdp':results[\"average\"][0]})\n",
    "    pdp_df.plot(x='grid_values', y='pdp', legend=None, ax=ax[row, col])\n",
    "\n",
    "fig.tight_layout()\n",
    "fig.savefig(Path(setup.temp_path).joinpath('71_PLOT_pdp_naics_onehot.png'),\n",
    "               bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "00f5a81f-7075-42bf-a660-88944a1298b5",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7687bacc-ac2c-4665-bcc0-58126b967dfa",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tf_p39",
   "language": "python",
   "name": "tf_p39"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}