65_xgb_dgi_mhier.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, DGI-Based Hierarchical Encoding\n",
    "Mean encode NAICS using DGI blending. Use levels similar to NAICS bas"
   ]
  },
  {
   "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": "9550b514-7063-4f3d-8a84-f36cf34c61a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import TargetEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f7051f7b-05fd-4ecb-8b9c-b8f3d896a684",
   "metadata": {},
   "outputs": [],
   "source": [
    "import statsmodels.api as sm\n",
    "from statsmodels.formula.api import ols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "fd09527b-32dd-47fa-b62f-f9307f89f553",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sklearn as sk\n",
    "from sklearn.inspection import permutation_importance, partial_dependence\n",
    "from scikitplot.metrics import plot_lift_curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0732ba40-8318-457b-b29f-f9b6de3b70aa",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sba_gnn.sba_gnn import sg_plot, sg_blender\n",
    "from sba_gnn.sba_gnn.sg_blender import HierarchicalEncoder # Target encoder"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d14a5947-6599-4253-acf3-91ddeffdedd3",
   "metadata": {},
   "source": [
    "## Import Processed Datasets"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85d31d6e-2d0c-4d95-b505-42173727446e",
   "metadata": {},
   "source": [
    "##### Get loans data, append clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f7fa98cc-da79-4996-8b23-aa0f3533bde1",
   "metadata": {},
   "outputs": [],
   "source": [
    "sba_loans = pd.read_parquet(Path(setup.temp_path).joinpath('01_DATA_transformed.parquet'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "32a6a5dd-0d35-48a0-9339-9a1ca11bdf41",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['NAICS', 'NAICS_2', 'NAICS_sector', 'NAICS_sector_desc', 'NAICS_num',\n",
       "       'NAICS_5', 'NAICS_4', 'NAICS_3', 'menc_NAICS', 'menc_NAICS_5',\n",
       "       'menc_NAICS_4', 'menc_NAICS_3', 'menc_NAICS_sector', 'mhier_NAICS',\n",
       "       'cenc_NAICS'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sba_loans.filter(like='NAICS').columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "63b6d02f-f147-42c9-92a8-e09e5978e029",
   "metadata": {},
   "outputs": [],
   "source": [
    "embed_df = pd.read_parquet(Path(setup.temp_path).joinpath('63_DATA_embeddings_tsne_naics.parquet'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e63cc797-deb1-4d66-945e-f184a2d9758d",
   "metadata": {},
   "source": [
    "##### NAICS levels to select NAICS-like clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "91345756-132f-40f0-8d3a-3b555dadd5d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[20.0, 106.0, 354.0, 834.0]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "naics_grp_stats = pd.read_csv(Path(setup.temp_path).joinpath('02_REPORT_naics_grp_stats_all.csv'))\n",
    "naics_grp_k = naics_grp_stats['count_grp'].drop_duplicates().sort_values().to_list()\n",
    "naics_grp_k"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "76727f09-d4f9-4654-925a-35bfdc3ce7b0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['cluster_003',\n",
       " 'cluster_010',\n",
       " 'cluster_020',\n",
       " 'cluster_106',\n",
       " 'cluster_354',\n",
       " 'cluster_834']"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cluster_cols = [c for c in embed_df if c.startswith('cluster_')]\n",
    "cluster_cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c4990eea-6247-4adf-ae56-c7f9d5668c96",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['cluster_834', 'cluster_354', 'cluster_106', 'cluster_020']"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cluster_cols_sel = sorted([c for c in cluster_cols if int(c.split('_')[1]) in naics_grp_k],\n",
    "                         reverse=True)\n",
    "cluster_cols_sel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "684d093a-3977-4263-a53d-92ffc0fabfff",
   "metadata": {},
   "outputs": [],
   "source": [
    "sba_loans = sba_loans.drop(columns = cluster_cols_sel, errors='ignore') \\\n",
    "    .merge(embed_df[['NAICS_orig'] + cluster_cols_sel] \\\n",
    "               .rename(columns={'NAICS_orig':'NAICS'}),\n",
    "           on='NAICS', how='left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "a8c6c341-9a49-4f08-818b-7c535dd48602",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "cluster_834\n",
       "800    27945\n",
       "154    19435\n",
       "198    14235\n",
       "372    13958\n",
       "16     12804\n",
       "8      12327\n",
       "7       9933\n",
       "378     9180\n",
       "584     8837\n",
       "169     8676\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sba_loans[cluster_cols_sel[0]].value_counts().head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b692da0-bc3e-452e-91bf-cee0fb496f55",
   "metadata": {},
   "source": [
    "## Fit the Hierarchical Target Encoder\n",
    "Also save it for later"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "73856b75-8bc5-40b0-ab01-8f08f7c4a3c3",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data = sba_loans[sba_loans['dset'] == 'train']\n",
    "hier_enc = HierarchicalEncoder()\n",
    "hier_enc.fit(train_data[cluster_cols_sel],train_data['target'])\n",
    "hier_col = hier_enc.transform(sba_loans[cluster_cols_sel]) \\\n",
    "        .rename('menc_dgi')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "c47fd06e-f3fd-4fdf-94a0-f3cae21044a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(Path(setup.temp_path).joinpath('65_DATA_encoder.pkl'), 'wb') as fout:\n",
    "    pickle.dump(hier_enc, fout)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b1210cac-bf19-4227-9fc5-a9c2f050bd92",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sba_gnn.sba_gnn.sg_blender.HierarchicalEncoder at 0x2807325e0>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hier_enc"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d56e9533-8071-494a-a7da-4917a03fb701",
   "metadata": {},
   "source": [
    "#### Set the predictor features.  Inlcude the mean-encoded NAICS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "add42230-18b9-4ccf-aaeb-5b33373fbf06",
   "metadata": {},
   "outputs": [],
   "source": [
    "sba_loans = pd.concat([sba_loans, hier_col], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "5c850a64-6c85-4d51-a75d-4c8ce23d4137",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['LoanNr_ChkDgt', 'Name', 'City', 'State', 'Zip', 'Bank', 'BankState',\n",
       "       'NAICS', 'ApprovalDate', 'ApprovalFY', 'Term', 'NoEmp', 'NewExist',\n",
       "       'CreateJob', 'RetainedJob', 'FranchiseCode', 'UrbanRural', 'RevLineCr',\n",
       "       'LowDoc', 'ChgOffDate', 'DisbursementDate', 'DisbursementGross',\n",
       "       'BalanceGross', 'MIS_Status', 'ChgOffPrinGr', 'GrAppv', 'SBA_Appv',\n",
       "       'target', 'guaranteed_fract', 'new_business', 'urban_flag',\n",
       "       'franchise_flag', 'NAICS_2', 'NAICS_sector', 'NAICS_sector_desc',\n",
       "       'NAICS_num', 'dset', 'dset_naics_holdout', 'NAICS_5', 'NAICS_4',\n",
       "       'NAICS_3', 'menc_NAICS', 'menc_NAICS_5', 'menc_NAICS_4', 'menc_NAICS_3',\n",
       "       'menc_NAICS_sector', 'mhier_NAICS',\n",
       "       'NS___Accommodation and Food Services',\n",
       "       'NS___Administrative and Support and Waste Management and Remediation Services',\n",
       "       'NS___Construction', 'NS___Health Care and Social Assistance',\n",
       "       'NS___Manufacturing',\n",
       "       'NS___Other Services (except Public Administration)',\n",
       "       'NS___Professional, Scientific, and Technical Services',\n",
       "       'NS___Retail Trade', 'NS___Wholesale Trade', 'NS___infrequent_sklearn',\n",
       "       'cenc_NAICS', 'cluster_834', 'cluster_354', 'cluster_106',\n",
       "       'cluster_020', 'menc_dgi'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sba_loans.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "6329f70e-9b92-483d-a235-909af8a55359",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save the new column\n",
    "sba_loans[['LoanNr_ChkDgt', 'dset', 'dset_naics_holdout', 'target', 'mhier_NAICS', 'menc_dgi',\n",
    "          'NAICS']] \\\n",
    "    .to_parquet(Path(setup.temp_path).joinpath('65_DATA_dgi_mhier.parquet'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "869a0f89-135f-46c1-a212-678987d1a490",
   "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>target</th>\n",
       "      <th>menc_dgi</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.222686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>menc_dgi</th>\n",
       "      <td>0.222686</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            target  menc_dgi\n",
       "target    1.000000  0.222686\n",
       "menc_dgi  0.222686  1.000000"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sba_loans[['target', 'menc_dgi']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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": 22,
   "id": "c07ddd4e-6987-454a-926a-3b59abf6caa6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['NoEmp', 'CreateJob', 'LowDoc', 'DisbursementGross', 'new_business', 'urban_flag', 'franchise_flag', 'menc_dgi']\n"
     ]
    }
   ],
   "source": [
    "predictor_features = setup.predictor_features + ['menc_dgi']\n",
    "print(predictor_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "1f8c475d-7ead-4f92-a19b-326343528f59",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(Path(setup.temp_path).joinpath('65_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": 24,
   "id": "af48634b-fc39-4c29-9792-bcf1cdf241c0",
   "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": 24,
     "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": "bc698909-3663-440f-98ce-fcfe312faa82",
   "metadata": {},
   "source": [
    "##### Fit using pre-determined parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "e8eca4a6-66e8-4eb2-b68a-1271f59cdbec",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_params = dict(best_params, \n",
    "                   **{'objective':\"binary:logistic\", 'n_estimators':setup.xgb_n_estimators})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "6d124de6-1d5d-4ff9-8071-4d05b0717f13",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_model = xgb.XGBClassifier(objective=\"binary:logistic\",\n",
    "                             n_estimators =  setup.xgb_n_estimators)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "9eebd329-7e39-4e52-be94-d839ad559b5d",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_model = xgb_model.set_params(**xgb_params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "a7f80910-9719-486b-8eac-93889064708a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0]\tvalidation_0-logloss:0.50049\tvalidation_1-logloss:0.50406\n",
      "[1]\tvalidation_0-logloss:0.49311\tvalidation_1-logloss:0.49693\n",
      "[2]\tvalidation_0-logloss:0.48691\tvalidation_1-logloss:0.49098\n",
      "[3]\tvalidation_0-logloss:0.48163\tvalidation_1-logloss:0.48592\n",
      "[4]\tvalidation_0-logloss:0.47721\tvalidation_1-logloss:0.48171\n",
      "[5]\tvalidation_0-logloss:0.47336\tvalidation_1-logloss:0.47805\n",
      "[6]\tvalidation_0-logloss:0.47009\tvalidation_1-logloss:0.47496\n",
      "[7]\tvalidation_0-logloss:0.46728\tvalidation_1-logloss:0.47230\n",
      "[8]\tvalidation_0-logloss:0.46481\tvalidation_1-logloss:0.46999\n",
      "[9]\tvalidation_0-logloss:0.46273\tvalidation_1-logloss:0.46807\n",
      "[10]\tvalidation_0-logloss:0.46091\tvalidation_1-logloss:0.46638\n",
      "[11]\tvalidation_0-logloss:0.45939\tvalidation_1-logloss:0.46500\n",
      "[12]\tvalidation_0-logloss:0.45808\tvalidation_1-logloss:0.46382\n",
      "[13]\tvalidation_0-logloss:0.45692\tvalidation_1-logloss:0.46280\n",
      "[14]\tvalidation_0-logloss:0.45593\tvalidation_1-logloss:0.46191\n",
      "[15]\tvalidation_0-logloss:0.45497\tvalidation_1-logloss:0.46109\n",
      "[16]\tvalidation_0-logloss:0.45417\tvalidation_1-logloss:0.46040\n",
      "[17]\tvalidation_0-logloss:0.45347\tvalidation_1-logloss:0.45979\n",
      "[18]\tvalidation_0-logloss:0.45288\tvalidation_1-logloss:0.45929\n",
      "[19]\tvalidation_0-logloss:0.45236\tvalidation_1-logloss:0.45890\n",
      "[20]\tvalidation_0-logloss:0.45193\tvalidation_1-logloss:0.45855\n",
      "[21]\tvalidation_0-logloss:0.45142\tvalidation_1-logloss:0.45816\n",
      "[22]\tvalidation_0-logloss:0.45104\tvalidation_1-logloss:0.45787\n",
      "[23]\tvalidation_0-logloss:0.45070\tvalidation_1-logloss:0.45764\n",
      "[24]\tvalidation_0-logloss:0.45042\tvalidation_1-logloss:0.45745\n",
      "[25]\tvalidation_0-logloss:0.45013\tvalidation_1-logloss:0.45724\n",
      "[26]\tvalidation_0-logloss:0.44989\tvalidation_1-logloss:0.45710\n",
      "[27]\tvalidation_0-logloss:0.44968\tvalidation_1-logloss:0.45694\n",
      "[28]\tvalidation_0-logloss:0.44948\tvalidation_1-logloss:0.45683\n",
      "[29]\tvalidation_0-logloss:0.44930\tvalidation_1-logloss:0.45670\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": 28,
     "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": "code",
   "execution_count": 29,
   "id": "9bbd8636-a5f1-4cbc-9498-329a78165472",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "target\n",
       "0    346866\n",
       "1     89254\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['target'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0913a102-a730-4b94-9977-e584df764d64",
   "metadata": {},
   "source": [
    "##### Save the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "af21dd7c-09d4-407f-9960-0748ba8f2247",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_model.save_model(Path(setup.temp_path).joinpath('65_MODEL_xgboost.json'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "8e5d3e7d-0bd5-4252-b287-c9cc65f7a132",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(Path(setup.temp_path).joinpath('65_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": 32,
   "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": 33,
   "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>0.183983</td>\n",
       "      <td>0.114966</td>\n",
       "      <td>0.016146</td>\n",
       "      <td>0.647492</td>\n",
       "      <td>0.167696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.281216</td>\n",
       "      <td>0.109653</td>\n",
       "      <td>0.016212</td>\n",
       "      <td>0.669335</td>\n",
       "      <td>0.283855</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            mean       std       min       max    median\n",
       "target                                                  \n",
       "0       0.183983  0.114966  0.016146  0.647492  0.167696\n",
       "1       0.281216  0.109653  0.016212  0.669335  0.283855"
      ]
     },
     "execution_count": 33,
     "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": 34,
   "id": "d7727498-825f-4f53-882f-cfcce730b1fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_train_df = pred_df[pred_df['dset'] == 'train']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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": 36,
   "id": "7a1cde01-a30b-475f-8fd7-409b13946c6d",
   "metadata": {},
   "outputs": [],
   "source": [
    "thresh_tune_data.to_csv(Path(setup.temp_path).joinpath('65_REPORT_thresh_tune.csv'), index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "947a487c-e71d-433c-bbf0-2ed7e57a908a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.24\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>12</th>\n",
       "      <td>0.24</td>\n",
       "      <td>0.452202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.22</td>\n",
       "      <td>0.451999</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    thresh        f1\n",
       "12    0.24  0.452202\n",
       "11    0.22  0.451999"
      ]
     },
     "execution_count": 37,
     "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": 38,
   "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": 39,
   "id": "8253c32c-0438-47b9-8570-7cde6f7f5590",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "predict_bin\n",
       "0    410067\n",
       "1    278014\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_df['predict_bin'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "dbb46f4a-a5ef-4fd6-9da4-30de4a00334d",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_df.to_parquet(Path(setup.temp_path).joinpath('65_DATA_combined_predictions.parquet'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ab860d0-eddc-4f63-b853-88da15d451f4",
   "metadata": {},
   "source": [
    "## Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>0.677844</td>\n",
       "      <td>0.428687</td>\n",
       "      <td>0.328408</td>\n",
       "      <td>0.617124</td>\n",
       "      <td>0.361704</td>\n",
       "      <td>0.722275</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>train</td>\n",
       "      <td>0.660545</td>\n",
       "      <td>0.452202</td>\n",
       "      <td>0.337597</td>\n",
       "      <td>0.684608</td>\n",
       "      <td>0.388384</td>\n",
       "      <td>0.733015</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>val</td>\n",
       "      <td>0.656954</td>\n",
       "      <td>0.447934</td>\n",
       "      <td>0.336237</td>\n",
       "      <td>0.670758</td>\n",
       "      <td>0.379415</td>\n",
       "      <td>0.723758</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>test</td>\n",
       "      <td>0.656337</td>\n",
       "      <td>0.448569</td>\n",
       "      <td>0.334777</td>\n",
       "      <td>0.679551</td>\n",
       "      <td>0.379871</td>\n",
       "      <td>0.726191</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>test</td>\n",
       "      <td>0.708741</td>\n",
       "      <td>0.391496</td>\n",
       "      <td>0.315542</td>\n",
       "      <td>0.515608</td>\n",
       "      <td>0.329468</td>\n",
       "      <td>0.714484</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        0.677844  0.428687         0.328408      0.617124   \n",
       "1  train        0.660545  0.452202         0.337597      0.684608   \n",
       "2    val        0.656954  0.447934         0.336237      0.670758   \n",
       "0   test        0.656337  0.448569         0.334777      0.679551   \n",
       "1   test        0.708741  0.391496         0.315542      0.515608   \n",
       "\n",
       "   average_precision_score  roc_auc_score  dset_naics_holdout  \n",
       "0                 0.361704       0.722275                 NaN  \n",
       "1                 0.388384       0.733015                 NaN  \n",
       "2                 0.379415       0.723758                 NaN  \n",
       "0                 0.379871       0.726191                 0.0  \n",
       "1                 0.329468       0.714484                 1.0  "
      ]
     },
     "execution_count": 41,
     "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('65_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": 42,
   "id": "82d9c0ab-8989-450d-a2b1-7997ca636052",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "pred_test_df = pred_df[pred_df['dset'] == 'test']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "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('65_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": 44,
   "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": 45,
   "id": "9586c637-7add-4401-b61c-37abd55b11b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "feat_impgain.to_csv(Path(setup.temp_path).joinpath('65_REPORT_importance_gain.csv'), index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>308.449524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LowDoc</th>\n",
       "      <td>126.420319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>menc_dgi</th>\n",
       "      <td>85.908852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DisbursementGross</th>\n",
       "      <td>48.965000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CreateJob</th>\n",
       "      <td>20.299299</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   importance\n",
       "urban_flag         308.449524\n",
       "LowDoc             126.420319\n",
       "menc_dgi            85.908852\n",
       "DisbursementGross   48.965000\n",
       "CreateJob           20.299299"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feat_impgain.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "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('65_PLOT_importance_gain.png'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27e5fb90-f305-40ff-acaf-285d7d056c15",
   "metadata": {},
   "source": [
    "##### Permutation Importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "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": 49,
   "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": 50,
   "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>7</th>\n",
       "      <td>menc_dgi</td>\n",
       "      <td>0.077711</td>\n",
       "      <td>0.077711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DisbursementGross</td>\n",
       "      <td>0.065936</td>\n",
       "      <td>0.065936</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>urban_flag</td>\n",
       "      <td>0.034164</td>\n",
       "      <td>0.034164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CreateJob</td>\n",
       "      <td>0.020811</td>\n",
       "      <td>0.020811</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LowDoc</td>\n",
       "      <td>0.014986</td>\n",
       "      <td>0.014986</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            features  importance  abs_importance\n",
       "7           menc_dgi    0.077711        0.077711\n",
       "3  DisbursementGross    0.065936        0.065936\n",
       "5         urban_flag    0.034164        0.034164\n",
       "1          CreateJob    0.020811        0.020811\n",
       "2             LowDoc    0.014986        0.014986"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "perm_importance_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "38059608-2f72-491f-86b3-0c1418b83531",
   "metadata": {},
   "outputs": [],
   "source": [
    "perm_importance_df \\\n",
    "    .to_csv(Path(setup.temp_path).joinpath('65_DATA_importance_perturb.csv'),\n",
    "            index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "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('65_PLOT_importance_perturb.png'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6fd44a7c-bcf5-4284-8b9f-b575fa523ff3",
   "metadata": {},
   "source": [
    "## PDP Plot - NAICS mean encoded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "0e9ee010-d1ce-4b1a-86ef-27d691583fc6",
   "metadata": {},
   "outputs": [],
   "source": [
    "results = partial_dependence(xgb_model, train_df[predictor_features], ['menc_dgi'])\n",
    "pdp_df = pd.DataFrame({'grid_values':results[\"grid_values\"][0],\n",
    "                       'pdp':results[\"average\"][0]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "730411c5-81d3-4911-ad8f-8743dc1617e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "3267866a-3453-443f-9015-5dbcd90e7594",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize = (8, 3))\n",
    "pdp_df.plot(x='grid_values', y='pdp', legend=None, ax=ax)\n",
    "ax.set_ylabel('effect')\n",
    "fig.savefig(Path(setup.temp_path).joinpath('65_PLOT_pdp_naics_meanenc.png'),\n",
    "               bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0a0ef4c8-fce6-43d5-b0c0-f354628157c0",
   "metadata": {},
   "source": [
    "## NAICS variation plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "3bd83c6f-be4f-4b38-a2af-e3f4923d434d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['cluster', 'cluster_003', 'cluster_010', 'cluster_020', 'cluster_106',\n",
       "       'cluster_354', 'cluster_834'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embed_df.filter(like='cluster').columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "aff8a20d-b664-4c09-b26e-d6a8d3ee400f",
   "metadata": {},
   "outputs": [],
   "source": [
    "sba_loans = sba_loans.drop(columns='cluster', errors='ignore') \\\n",
    "    .merge(embed_df[['NAICS_orig', 'cluster']] \\\n",
    "               .rename(columns={'NAICS_orig':'NAICS'}),\n",
    "           on='NAICS', how='left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "74a87f6b-05af-42dc-80be-cc8f0ddac871",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['cluster_834', 'cluster_354', 'cluster_106', 'cluster_020', 'cluster'], dtype='object')"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sba_loans.filter(like='cluster').columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "4686a4a5-0d81-4f86-a8da-f4f44a2a416a",
   "metadata": {},
   "outputs": [],
   "source": [
    "sba_loans['cluster_desc'] = ''\n",
    "naics_info = sg_plot.naics_variance_data(sba_loans,\n",
    "                                        naics_sector_feat='cluster',\n",
    "                                        naics_sector_desc_feat='cluster_desc')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "11125d4e-4b73-4c69-914d-9b46b6e868b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "sba_loans['cluster_desc'] = ''\n",
    "naics_info_020 = sg_plot.naics_variance_data(sba_loans,\n",
    "                                        naics_sector_feat='cluster_020',\n",
    "                                        naics_sector_desc_feat='cluster_desc')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "52c2da4e-6c97-4fef-8821-937ab0558287",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Plot higher volume NAICS (k=10 cluster)\n",
    "plt.close()\n",
    "fig = sg_plot.naics_variance_plot(naics_info[naics_info['count_naics'] >= 100],\n",
    "                                  naics_sector_feat='cluster',\n",
    "                                  naics_sector_desc_feat='cluster_desc',\n",
    "                                  num_sectors=5, xlabel='loan default rate')\n",
    "fig.savefig(Path(setup.temp_path).joinpath('65_PLOT_naics_counts_rates_100.png'),\n",
    "           bbox_inches='tight')\n",
    "plt.close()\n",
    "fig = sg_plot.naics_variance_plot(naics_info[naics_info['count_naics'] >= 150],\n",
    "                                  naics_sector_feat='cluster',\n",
    "                                  naics_sector_desc_feat='cluster_desc',\n",
    "                                  num_sectors=5, xlabel='loan default rate')\n",
    "fig.savefig(Path(setup.temp_path).joinpath('65_PLOT_naics_counts_rates_150.png'),\n",
    "           bbox_inches='tight')\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "952e5583-252f-4d2e-b925-468ee76f5b3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Plot higher volume NAICS, 20 level cluster\n",
    "plt.close()\n",
    "fig = sg_plot.naics_variance_plot(naics_info_020[naics_info_020['count_naics'] >= 100],\n",
    "                                  naics_sector_feat='cluster_020',\n",
    "                                  naics_sector_desc_feat='cluster_desc',\n",
    "                                  num_sectors=5, xlabel='loan default rate')\n",
    "fig.savefig(Path(setup.temp_path).joinpath('65_PLOT_naics_counts_rates_clus020_100.png'),\n",
    "           bbox_inches='tight')\n",
    "plt.close()\n",
    "fig = sg_plot.naics_variance_plot(naics_info_020[naics_info_020['count_naics'] >= 150],\n",
    "                                  naics_sector_feat='cluster_020',\n",
    "                                  naics_sector_desc_feat='cluster_desc',\n",
    "                                  num_sectors=5, xlabel='loan default rate')\n",
    "fig.savefig(Path(setup.temp_path).joinpath('65_PLOT_naics_counts_rates_clus020_150.png'),\n",
    "           bbox_inches='tight')\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "a68836ab-3cd4-480b-b878-5dece3ada8a0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x800 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot top 8 volume NAICS, 20 level cluster\n",
    "plt.close()\n",
    "fig = sg_plot.naics_variance_plot(naics_info_020[naics_info_020['count_naics'] >= 100],\n",
    "                                  naics_sector_feat='cluster_020',\n",
    "                                  naics_sector_desc_feat='cluster_desc',\n",
    "                                  num_sectors=8, xlabel='loan default rate')\n",
    "fig.savefig(Path(setup.temp_path).joinpath('65_PLOT_naics_counts_rates_clus020_100_top_8.png'),\n",
    "           bbox_inches='tight')\n",
    "plt.close()\n",
    "fig = sg_plot.naics_variance_plot(naics_info_020[naics_info_020['count_naics'] >= 150],\n",
    "                                  naics_sector_feat='cluster_020',\n",
    "                                  naics_sector_desc_feat='cluster_desc',\n",
    "                                  num_sectors=8, xlabel='loan default rate')\n",
    "fig.savefig(Path(setup.temp_path).joinpath('65_PLOT_naics_counts_rates_clus020_150_top_8.png'),\n",
    "           bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4c280c44-14dc-4c38-b4ad-caa01a2aabc2",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "77c0754b-5349-495b-9536-772a449f73bd",
   "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
}