04_xgb_menc.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 Target Encoded NAICS\n",
    "Same as the baseline but include the target (mean) encoded NAICS"
   ]
  },
  {
   "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": "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": "31bf325d-4e9b-4a98-9927-83f6ad19fc11",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = sba_loans[sba_loans['dset'] == 'train']\n",
    "test_df = sba_loans[sba_loans['dset'] == 'test']\n",
    "val_df = sba_loans[sba_loans['dset'] == 'val']"
   ]
  },
  {
   "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": 8,
   "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_NAICS']\n"
     ]
    }
   ],
   "source": [
    "predictor_features = setup.predictor_features + ['menc_NAICS']\n",
    "print(predictor_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "1f8c475d-7ead-4f92-a19b-326343528f59",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(Path(setup.temp_path).joinpath('04_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": 10,
   "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": 10,
     "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": 11,
   "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": 12,
   "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": 13,
   "id": "9eebd329-7e39-4e52-be94-d839ad559b5d",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_model = xgb_model.set_params(**xgb_params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a7f80910-9719-486b-8eac-93889064708a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0]\tvalidation_0-logloss:0.50028\tvalidation_1-logloss:0.50383\n",
      "[1]\tvalidation_0-logloss:0.49273\tvalidation_1-logloss:0.49654\n",
      "[2]\tvalidation_0-logloss:0.48634\tvalidation_1-logloss:0.49040\n",
      "[3]\tvalidation_0-logloss:0.48095\tvalidation_1-logloss:0.48523\n",
      "[4]\tvalidation_0-logloss:0.47645\tvalidation_1-logloss:0.48094\n",
      "[5]\tvalidation_0-logloss:0.47250\tvalidation_1-logloss:0.47715\n",
      "[6]\tvalidation_0-logloss:0.46919\tvalidation_1-logloss:0.47397\n",
      "[7]\tvalidation_0-logloss:0.46633\tvalidation_1-logloss:0.47126\n",
      "[8]\tvalidation_0-logloss:0.46381\tvalidation_1-logloss:0.46888\n",
      "[9]\tvalidation_0-logloss:0.46169\tvalidation_1-logloss:0.46692\n",
      "[10]\tvalidation_0-logloss:0.45978\tvalidation_1-logloss:0.46513\n",
      "[11]\tvalidation_0-logloss:0.45824\tvalidation_1-logloss:0.46371\n",
      "[12]\tvalidation_0-logloss:0.45689\tvalidation_1-logloss:0.46252\n",
      "[13]\tvalidation_0-logloss:0.45571\tvalidation_1-logloss:0.46146\n",
      "[14]\tvalidation_0-logloss:0.45464\tvalidation_1-logloss:0.46050\n",
      "[15]\tvalidation_0-logloss:0.45372\tvalidation_1-logloss:0.45968\n",
      "[16]\tvalidation_0-logloss:0.45295\tvalidation_1-logloss:0.45903\n",
      "[17]\tvalidation_0-logloss:0.45224\tvalidation_1-logloss:0.45843\n",
      "[18]\tvalidation_0-logloss:0.45158\tvalidation_1-logloss:0.45787\n",
      "[19]\tvalidation_0-logloss:0.45103\tvalidation_1-logloss:0.45743\n",
      "[20]\tvalidation_0-logloss:0.45059\tvalidation_1-logloss:0.45708\n",
      "[21]\tvalidation_0-logloss:0.45008\tvalidation_1-logloss:0.45670\n",
      "[22]\tvalidation_0-logloss:0.44965\tvalidation_1-logloss:0.45637\n",
      "[23]\tvalidation_0-logloss:0.44932\tvalidation_1-logloss:0.45612\n",
      "[24]\tvalidation_0-logloss:0.44903\tvalidation_1-logloss:0.45590\n",
      "[25]\tvalidation_0-logloss:0.44880\tvalidation_1-logloss:0.45573\n",
      "[26]\tvalidation_0-logloss:0.44845\tvalidation_1-logloss:0.45548\n",
      "[27]\tvalidation_0-logloss:0.44827\tvalidation_1-logloss:0.45536\n",
      "[28]\tvalidation_0-logloss:0.44805\tvalidation_1-logloss:0.45525\n",
      "[29]\tvalidation_0-logloss:0.44785\tvalidation_1-logloss:0.45512\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": 14,
     "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": 15,
   "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": 15,
     "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": 16,
   "id": "af21dd7c-09d4-407f-9960-0748ba8f2247",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_model.save_model(Path(setup.temp_path).joinpath('04_MODEL_xgboost.json'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "8e5d3e7d-0bd5-4252-b287-c9cc65f7a132",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(Path(setup.temp_path).joinpath('04_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": 18,
   "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": 19,
   "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>0.185065</td>\n",
       "      <td>0.114651</td>\n",
       "      <td>0.015973</td>\n",
       "      <td>0.652089</td>\n",
       "      <td>0.170870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.284072</td>\n",
       "      <td>0.110439</td>\n",
       "      <td>0.016630</td>\n",
       "      <td>0.678520</td>\n",
       "      <td>0.286289</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            mean       std       min       max    median\n",
       "target                                                  \n",
       "0       0.185065  0.114651  0.015973  0.652089  0.170870\n",
       "1       0.284072  0.110439  0.016630  0.678520  0.286289"
      ]
     },
     "execution_count": 19,
     "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": 20,
   "id": "d7727498-825f-4f53-882f-cfcce730b1fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_train_df = pred_df[pred_df['dset'] == 'train']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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": 22,
   "id": "7a1cde01-a30b-475f-8fd7-409b13946c6d",
   "metadata": {},
   "outputs": [],
   "source": [
    "thresh_tune_data.to_csv(Path(setup.temp_path).joinpath('04_REPORT_thresh_tune.csv'), index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>0.454738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.22</td>\n",
       "      <td>0.454412</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    thresh        f1\n",
       "12    0.24  0.454738\n",
       "11    0.22  0.454412"
      ]
     },
     "execution_count": 23,
     "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": 24,
   "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": 25,
   "id": "8253c32c-0438-47b9-8570-7cde6f7f5590",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "predict_bin\n",
       "0    406604\n",
       "1    281477\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_df['predict_bin'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "dbb46f4a-a5ef-4fd6-9da4-30de4a00334d",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_df.to_parquet(Path(setup.temp_path).joinpath('04_DATA_combined_predictions.parquet'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ab860d0-eddc-4f63-b853-88da15d451f4",
   "metadata": {},
   "source": [
    "## Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>0.658387</td>\n",
       "      <td>0.437470</td>\n",
       "      <td>0.322860</td>\n",
       "      <td>0.678231</td>\n",
       "      <td>0.368420</td>\n",
       "      <td>0.728298</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>train</td>\n",
       "      <td>0.665700</td>\n",
       "      <td>0.454738</td>\n",
       "      <td>0.341294</td>\n",
       "      <td>0.681146</td>\n",
       "      <td>0.393421</td>\n",
       "      <td>0.735795</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>val</td>\n",
       "      <td>0.661459</td>\n",
       "      <td>0.449984</td>\n",
       "      <td>0.339400</td>\n",
       "      <td>0.667457</td>\n",
       "      <td>0.386203</td>\n",
       "      <td>0.727009</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>test</td>\n",
       "      <td>0.660628</td>\n",
       "      <td>0.449299</td>\n",
       "      <td>0.337199</td>\n",
       "      <td>0.673048</td>\n",
       "      <td>0.383669</td>\n",
       "      <td>0.728512</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>test</td>\n",
       "      <td>0.655168</td>\n",
       "      <td>0.419852</td>\n",
       "      <td>0.302365</td>\n",
       "      <td>0.686659</td>\n",
       "      <td>0.347722</td>\n",
       "      <td>0.729576</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        0.658387  0.437470         0.322860      0.678231   \n",
       "1  train        0.665700  0.454738         0.341294      0.681146   \n",
       "2    val        0.661459  0.449984         0.339400      0.667457   \n",
       "0   test        0.660628  0.449299         0.337199      0.673048   \n",
       "1   test        0.655168  0.419852         0.302365      0.686659   \n",
       "\n",
       "   average_precision_score  roc_auc_score  dset_naics_holdout  \n",
       "0                 0.368420       0.728298                 NaN  \n",
       "1                 0.393421       0.735795                 NaN  \n",
       "2                 0.386203       0.727009                 NaN  \n",
       "0                 0.383669       0.728512                 0.0  \n",
       "1                 0.347722       0.729576                 1.0  "
      ]
     },
     "execution_count": 27,
     "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('04_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": 28,
   "id": "82d9c0ab-8989-450d-a2b1-7997ca636052",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "pred_test_df = pred_df[pred_df['dset'] == 'test']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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('04_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": 30,
   "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": 31,
   "id": "9586c637-7add-4401-b61c-37abd55b11b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "feat_impgain.to_csv(Path(setup.temp_path).joinpath('04_REPORT_importance_gain.csv'), index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>275.510254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LowDoc</th>\n",
       "      <td>120.972610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>menc_NAICS</th>\n",
       "      <td>97.928482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DisbursementGross</th>\n",
       "      <td>47.278519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>franchise_flag</th>\n",
       "      <td>20.430019</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   importance\n",
       "urban_flag         275.510254\n",
       "LowDoc             120.972610\n",
       "menc_NAICS          97.928482\n",
       "DisbursementGross   47.278519\n",
       "franchise_flag      20.430019"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feat_impgain.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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('04_PLOT_importance_gain.png'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27e5fb90-f305-40ff-acaf-285d7d056c15",
   "metadata": {},
   "source": [
    "##### Permutation Importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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": 35,
   "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": 36,
   "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_NAICS</td>\n",
       "      <td>0.086456</td>\n",
       "      <td>0.086456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DisbursementGross</td>\n",
       "      <td>0.065399</td>\n",
       "      <td>0.065399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>urban_flag</td>\n",
       "      <td>0.034239</td>\n",
       "      <td>0.034239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CreateJob</td>\n",
       "      <td>0.019439</td>\n",
       "      <td>0.019439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LowDoc</td>\n",
       "      <td>0.014674</td>\n",
       "      <td>0.014674</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            features  importance  abs_importance\n",
       "7         menc_NAICS    0.086456        0.086456\n",
       "3  DisbursementGross    0.065399        0.065399\n",
       "5         urban_flag    0.034239        0.034239\n",
       "1          CreateJob    0.019439        0.019439\n",
       "2             LowDoc    0.014674        0.014674"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "perm_importance_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "38059608-2f72-491f-86b3-0c1418b83531",
   "metadata": {},
   "outputs": [],
   "source": [
    "perm_importance_df \\\n",
    "    .to_csv(Path(setup.temp_path).joinpath('04_DATA_importance_perturb.csv'),\n",
    "            index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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('04_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": 39,
   "id": "0e9ee010-d1ce-4b1a-86ef-27d691583fc6",
   "metadata": {},
   "outputs": [],
   "source": [
    "results = partial_dependence(xgb_model, train_df[predictor_features], ['menc_NAICS'])\n",
    "pdp_df = pd.DataFrame({'grid_values':results[\"grid_values\"][0],\n",
    "                       'pdp':results[\"average\"][0]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "730411c5-81d3-4911-ad8f-8743dc1617e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "3267866a-3453-443f-9015-5dbcd90e7594",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArwAAAEmCAYAAACAr0P7AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAA9hAAAPYQGoP6dpAABG0klEQVR4nO3deXhTVf4/8HeWJumWdKU7XWjLTguUVhRQpLKoCAMqMowgbuM6XyzIMiq4FxFHVBBm+I0yOiK4oOMMCA6VumBZLGUtlLYUuqZ0oUnXJE3O749KtNMCXdKmTd+v58ljc3PvyeeeJ6RvT889VyKEECAiIiIiclBSexdARERERNSVGHiJiIiIyKEx8BIRERGRQ2PgJSIiIiKHxsBLRERERA6NgZeIiIiIHBoDLxERERE5NAZeIiIiInJocnsX0BNZLBYUFxfD3d0dEonE3uUQERER0f8QQqC6uhqBgYGQSq8+hsvA24ri4mKEhITYuwwiIiIiuoaCggIEBwdfdR8G3la4u7sDaOpAtVpt52qIiIiI6H/p9XqEhIRYc9vV9IjAu2HDBrz++uvQarWIiYnBO++8g/j4+Fb33bFjB1599VXk5OTAZDIhKioKixcvxr333mvdRwiBVatWYfPmzaiqqsINN9yAjRs3Iioqqk31XJ7GoFarGXiJiIiIerC2TD+1+0Vr27dvR1JSElatWoUjR44gJiYGU6ZMwcWLF1vd38vLC8888wzS0tJw/PhxLFy4EAsXLsSePXus+6xZswZvv/02Nm3ahIMHD8LV1RVTpkxBQ0NDd50WEREREfUQEiGEsGcBCQkJGDNmDNavXw+g6YKxkJAQPPnkk1i+fHmb2hg1ahRuu+02vPTSSxBCIDAwEIsXL8aSJUsAADqdDn5+ftiyZQvuueeea7an1+uh0Wig0+k4wktERETUA7Unr9l1hNdoNCI9PR2JiYnWbVKpFImJiUhLS7vm8UIIpKSkICsrCxMmTAAA5OXlQavVNmtTo9EgISHhim0aDAbo9fpmDyIiIiJyDHYNvOXl5TCbzfDz82u23c/PD1qt9orH6XQ6uLm5QaFQ4LbbbsM777yDW265BQCsx7WnzeTkZGg0GuuDKzQQEREROQ67z+HtCHd3dxw9ehSHDx/GK6+8gqSkJKSmpna4vRUrVkCn01kfBQUFtiuWiIiIiOzKrqs0+Pj4QCaTobS0tNn20tJS+Pv7X/E4qVSKyMhIAEBsbCxOnz6N5ORk3HTTTdbjSktLERAQ0KzN2NjYVttTKpVQKpWdPBsiIiKivqvW0IhaYyP6uavsXUoLdh3hVSgUGD16NFJSUqzbLBYLUlJSMHbs2Da3Y7FYYDAYAADh4eHw9/dv1qZer8fBgwfb1SYRERERXZsQArtPapH4l++w9LPjsPN6CK2y+zq8SUlJWLBgAeLi4hAfH49169ahtrYWCxcuBADMnz8fQUFBSE5OBtA03zYuLg4DBgyAwWDArl278OGHH2Ljxo0AmtZiW7RoEV5++WVERUUhPDwczz33HAIDAzFz5kx7nSYRERGRwym8VIdV/zqFlDNNy8nKZRKU1xjh696z/nJu98A7Z84clJWVYeXKldBqtYiNjcXu3butF53l5+c3uz9ybW0tHnvsMRQWFsLZ2RmDBg3CP//5T8yZM8e6z9KlS1FbW4uHH34YVVVVGDduHHbv3g2VqucNsRMRERHZ0/nyWry59yzKqg0tXpPLpBgSoMboUE+M6u8Bb7emIGsyW/D3H/Pw1t5s1JvMcJJJ8PCECDwxMQrOCll3n8I12X0d3p6I6/ASERGRo7NYBP558AKSd51BvcncpmMifFwxKtQTJwp1yCqtBgDEh3vhlZnDEOV37Vv82lJ78prdR3iJiIiIqHsVV9Vj6WfH8WNOOQDg+gHemDMmpMVtemsNjThWUIX0C5eQfbEG58prca68FgDg5arAn28djNmjgtp0e197YuAlIiIi6iOEENhxpAjP//sUqhsaoXKSYsW0wbj3ulBIpa2H1rnx/QEAVXVGZOQ3hV+5TIIFY8Pg6arozvI7jIGXiIiIyMF8nl6Iv31/Dkazpdl2Y6MFRVX1AICR/T3wxl0xiPB1a1ObHi4KTBzUDxMH9bN5vV2NgZeIiIjIgXxyuABLPz9+xdedZBIsSozGHydEQC7rlfcgazcGXiIiIiIH8a+jRVi2oynszh8bijtiAlvsE+LlAj9131q5ioGXiIiIyAHsOlGCpE+OQQjgD9f1xwt3DO3xF5N1l74xjk1ERETkwPZmluJPH2fAbBG4a3QwXrxjGMPubzDwEhEREfVi350tw2MfHUGjRWBGbCBWzx5xxRUX+ipOaSAiIiLqARrNFhwr1OGH7DKcL69FoIczQr1dEOLlglBvV/irVTCZLcjSViOzRI/MYj0yS/Q4XlgFk1lg2jB/vHFXDGQMuy0w8BIREVGf0mAy48fs8hZLdgGA2SLQaLHAZBZNP5ubflY6SeGmlEOtcoKbSg53lRyuCnmnw2Wd0YxDeZX4IbsM+3PKoW9ovOK+CpkUjRYLLK3cI3fKUD+8dc/IPrPqQnsx8BIREVGfkrzrNP6RdsHeZbRKrZJjXJQPhgZqoNU1IL+yDvmVdSi8VGcN6N6uCgwJVDc9Apoekf3cOGf3Khh4iYiIqM+objDh0/RCAEBsiAcU8uYjolIJ4CSTQiaVQC6VwkkmgUwqgaHRgpqGRtQYGlHdYPrlv41oZbC1XaQSYHiQBuOjfDE+ygcjgj1aHTU2WwRKdPVQyKTwdVcy3LYTAy8RERH1GV8eLUad0YwIX1d88dj1vSY4yqQSBHu62LuMXosTPYiIiKhPEELgowNNUxnmJYT2mrBLncfAS0RERH3CkfwqnNFWQymX4s5RwfYuh7oRAy8RERH1CZdHd6fHBELj4mTnaqg7MfASERGRw7tUa8R/TpQAAOYl9LdzNdTdGHiJiIjI4X1+pBDGRguGBKgRG+Jh73KomzHwEhERkUMTQuCjg/kAgD9cx4vV+iIGXiIiInJoP+VWIK+8Fm5KOe6IDbR3OWQHDLxERETk0D462HSx2syRgXBT8hYEfREDLxERETmsi/oGfHOqFEDT2rvUNzHwEhERkcP65OcCNFoERod6YnCA2t7lkJ0w8BIREZFDMlsEPj5UAIBLkfV1DLxERETkcHIuVmPu5gMoqqqHh4sTbh0eYO+SyI44c5uIiIgcRoPJjHf35WDjd7kwmQWcnWR4ccYwqJxk9i6N7IiBl4iIiBzCj9nlePbLEzhfUQcAuHlQP7w4YyiCPV3sXBnZGwMvERER9UhCCFTWGnGuvBZ5ZbXILa9BXlkttPoGCNF8X5PZgjPaagCAn1qJ56cPxdRh/rzJBAFg4CUiIqIe6Kecciz+9BhKdA1tPkYiARaMDcPiydFwVzl1YXXU2zDwEhERUY/y7ZlSPPLPIzA2WiCRAIEaZ0T4uiLCxxXhPq4I9nSBTNZy5DbMu+l1ov/FwEtEREQ9xtcnSvCnbRkwmQVuGeKHdXNi4cq7o1En8RNEREREPcIXGYVY/MkxWAQwPSYQf7k7Bk4yrqBKncfAS0RERHa39WA+nvnyBIQA7hodjNWzR0Am5QVnZBsMvERERGRX7/2Yhxf/kwkAmD82FM9PHwopwy7ZEAMvERER2c3ff8zDS7+E3T9OiMDyaYO4lBjZHAMvERER2cUHaeetYfeJiZFYPDmaYZe6BGeCExERUbfbejAfK/91CgDwyI0DGHapSzHwEhERUbf65HAB/vzFCQDAg+PCsWzqQIZd6lIMvERERNRtPk8vxLIdxwEA910fhmduG8ywS12uRwTeDRs2ICwsDCqVCgkJCTh06NAV9928eTPGjx8PT09PeHp6IjExscX+9913HyQSSbPH1KlTu/o0iIiI6Aoqa41Y/202nv7sGIQA7r0uFKumD2HYpW5h94vWtm/fjqSkJGzatAkJCQlYt24dpkyZgqysLPTr16/F/qmpqZg7dy6uv/56qFQqvPbaa5g8eTJOnTqFoKAg635Tp07F+++/b32uVCq75XyIiIioiRAC6Rcu4aOD+dh5ogTGRgsAYG58CF64YyjDLnUbiRBC2LOAhIQEjBkzBuvXrwcAWCwWhISE4Mknn8Ty5cuvebzZbIanpyfWr1+P+fPnA2ga4a2qqsKXX37ZoZr0ej00Gg10Oh3UanWH2iAiInIEDSYzCi/Vo6CyDhW1xjYfd6nWiM/SC5FVWm3dNjRQjQVjw3Dn6GCus0ud1p68ZtcRXqPRiPT0dKxYscK6TSqVIjExEWlpaW1qo66uDiaTCV5eXs22p6amol+/fvD09MTNN9+Ml19+Gd7e3q22YTAYYDAYrM/1en0HzoaIiKh3K7xUh59yKpB+4RLyKmpRUFkHrb4BnRkaUzlJMSMmCL9P6I8RwRqO6pJd2DXwlpeXw2w2w8/Pr9l2Pz8/nDlzpk1tLFu2DIGBgUhMTLRumzp1KmbNmoXw8HDk5ubiz3/+M6ZNm4a0tDTIZLIWbSQnJ+OFF17o3MkQERH1IhaLQFmNAT+fv4T9ueX4Kacc5yvqWt3XVSFDf29X+Lor0daBWblUgvFRvpg5MggaZycbVk7Ufnafw9sZq1evxrZt25CamgqVSmXdfs8991h/Hj58OEaMGIEBAwYgNTUVkyZNatHOihUrkJSUZH2u1+sREhLStcUTERF1ISEESvUG5JbV4FxZDfIr61Cia0CpvsH6X5O5+dCtTCpBTLAGYwd4I9rPHf29XNDfywVergqOzFKvZtfA6+PjA5lMhtLS0mbbS0tL4e/vf9Vj165di9WrV2Pv3r0YMWLEVfeNiIiAj48PcnJyWg28SqWSF7UREVGv1mi2YO/pi/jmlBY5ZTXIvViDWqP5qsdIJEB0P3dcH+mNGwb4ICHCC+4qjsaS47Fr4FUoFBg9ejRSUlIwc+ZMAE0XraWkpOCJJ5644nFr1qzBK6+8gj179iAuLu6a71NYWIiKigoEBATYqnQiIqIeoVTfgG2HCvDxoXxo9Q3NXpNJJQj1ckGErytCvV0RoFEhQOMMf40KARoVfN2VcJL1iBVKibqU3ac0JCUlYcGCBYiLi0N8fDzWrVuH2tpaLFy4EAAwf/58BAUFITk5GQDw2muvYeXKldi6dSvCwsKg1WoBAG5ubnBzc0NNTQ1eeOEFzJ49G/7+/sjNzcXSpUsRGRmJKVOm2O08iYiIbKXO2Iifz1/CtsP5+OZUKRotTVMTvF0VuHN0MEaFemKAryv6e7lCIWegJbJ74J0zZw7KysqwcuVKaLVaxMbGYvfu3dYL2fLz8yGV/vqPdePGjTAajbjzzjubtbNq1So8//zzkMlkOH78OP7xj3+gqqoKgYGBmDx5Ml566SVOWyAiol7HYhE4V16LowVVyMi/hKMFVTijrYbZ8uv82zFhnvjDdaGYOswfSnnLi7OJ+jq7r8PbE3EdXiIisodGswW5ZbU4VazDySI9ThXrkFmsR7WhscW+fmolbhnihz9cF4pB/vxdRX1Pr1mHl4iIqK85eK4Cz3x5EuU1hhav1RnN1ruR/ZZSLsWIYA1G9vdEbIgHRvb3QIDGuTvKJXIIDLxERETd5FBeJRZuOYy6q6ye4KqQYWigBkMC1RgaqMawIA0i+7nx4jKiTmDgJSIi6gY/n6/EwvcPoc5oxvgoH6y8fUiLtW2VcimCPJx5210iG2PgJSIi6mJH8i/hvvcPo9Zoxg2R3tg8Pw4qJ15cRtRd+PcRIiKiLnS0oAoL/n4INYZGjI3wxv+bP4Zhl6ibcYSXiIiok+qNZpTXGCAEYBHilwdQXFWPx7ceQbWhEQnhXvj7fXFwVjDsEnU3Bl4iIqJWpF+oRHZpTYvtjRaBEl09CirrUXCpDgWV9a2uuPBb8WFeeO++MXBR8NcukT3wXx4REdH/+Ot3uUj++ky7jlHIpZBLJZBKJJBIYP3vdeHeWHt3DFyV/JVLZC/810dERPQbG/bl4PU9WQCAsRHeLYKqVAL4qVUI8XJGiKcLgj1dEOLlDI2zU4tVF4ioZ2DgJSIi+sVbe7Px5t6zAICkW6Lxp0lRdq6IiGyBgZeIiOyiwWSGvsEETxeF3W+qIITAm/89i7e/zQEALJ06EI/dFGnXmojIdhh4iYio25XXGPC7d/ejoLIeAODh4gQvVwV8XJXwcVdg6rAATB8R0C1TBIQQWLMnCxtTcwEAz9w6GA9NiOjy9yWi7sPAS0RE3cpktuDxj45Ywy4AVNWZUFVnwrmyWgDArhNafPpzAV6eOQyh3q5dUoeh0Ywfzpbjs/RC7D6lBQCsvH0I7h8X3iXvR0T2w8BLRETdKnnXGRzMq4SrQoYdj90AHzcFKmqNqKgxoqLWgMxiPf7fj3n4Ibsck9/8Hn+aFIWHxkdAIe/8tIfLIXfXiRL8N7MU1YZG62svzhiK+WPDOv0eRNTzSIQQwt5F9DR6vR4ajQY6nQ5qtdre5RAROYwvMgrx1PZjAIBNfxiNqcP8W93vfHktnvnyBPbnVAAABvq549VZwzA61Kvd7ymEwM8XLuGznwux60RJs5Drp1bi1uEB+N3IIIwI9mj/CRGR3bQnrzHwtoKBl4jI9k4W6TB7408wNFrwxMRILJky8Kr7CyHw5dEivPSf06isNQIAfp/QH8umDoLG2ema71dQWYcdR4qwI6MQFyrqrNsvh9zbhgdgVH9PSKVcSoyoN2Lg7SQGXiIi26qsNWL6Oz+iqKoeNw30xd8XjIGsjUHzUq0Rr+46jU/TCwEAvu5KrJo+BLcNb3lRm6HRjN0ntfj4UD4OnKu0bndVyHDr8ADMHh2M+DAvhlwiB8DA20kMvEREttNotuC+9w/jx5xyhHq74KvHx0Hjcu0R2v914FwF/vzFCeuFbRMH+uLFGcMQ4uWCc2U1+PhQPj5LL8SlOhMAQCIBrh/gjdmjgjF1mD9v60vkYBh4O4mBl4io8xpMZuw5pcXWg/k4mFcJZycZvnj8egzy7/j3qqHRjI2puXh3Xy6MZgucnWQYEqhG+oVL1n381SrMGROCu8eEIMjD2RanQkQ9UHvyGv93l4iIbEYIgYyCKnyWXoh/HytGdUPTBWIyqQRr74rpVNgFAKVchkWJ0bh9RCCe+eIEDuZVIv3CJUglwMSB/TA3vj9uGugLuZ1vZEFEPQsDLxERdVpRVT2+zCjCjiOFyP1lygEABHk4Y/boYNw5Khj9vV1s9n6R/dyw7eHr8NWxYhRV1WNmbBACOZpLRFfAwEtERB1S3WDC1ye1+OJIEdLOVVi3q5ykuHVYAO4cHYzrIry77AIxiUSCGbFBXdI2ETkWBl4iImqz8hoDvssqw7dnLiLlTCkaTBbra9dFeGHWyGBMG+4Pd1X7L0ojIuoqDLxERHRFQggcK9Rh35mLSM26iONFOvz2UucBvq6YNSoYM2IDEexpuykLRES2xMBLREStEkLg/7YdxVfHipttHxakxsSB/ZA42A8jgjUt1sIlIuppGHiJiKhVXx4twlfHiiGXSjB5qB9uGtgPN0X7op9aZe/SiIjahYGXiIhauKhvwPNfZQIAnrolGo9PjLRzRUREHceFComIqBkhBP78xUno6k0YHqTBHydE2LskIqJOYeAlIqJmvjpWjL2nS+Ekk+D1u0bwJg5E1OvxW4yIiKwuVjdg1VenAABP3hzV6TujERH1BJzDS0TkYIQQqDOaUVlrRIPJDHeVE9TOcjg7ya66ooIQAs9+cRJVdSYMCVDj0ZsGdGPVRERdh4GXiKiXyC6txqMfHUHhpTq4KeVwVcrhqpDDTSmH0kmKqjoTKmuNqKg1NLshxGVyqQRqZyeoVXKEersiLtQTo8M8ERviAReFHP8+XoJvMkshl0qw9q4YOHEqAxE5CAZeIqJeoLzGgIVbDqPwUj0AoMFkRHmN8arHKOVSqJxkqDE0wmwRaLQIVNYaUVlrxPmKOnx3tgwAIJNKMCRAjfzKOgDAEzdHYkggpzIQkeNg4CUi6uEaTGY8/MHPKLxUj1BvF/zt3jhYhECdsRE1BjNqDY1oMJmhcXaCl6sCPm5KeLkq4KJomsJweYqDvsGE6oZGVNWZcLpEj58vXEL6+UoU6xpwokgHABgcoMZjN3EJMiJyLAy8REQ9mBACyz4/jiP5VVCr5HjvvjEY4OvWrjYkEknT9AelHAGapm3x4V5YcH0YAKC4qh4/X7iEnNJq3BUXAoWcUxmIyLF06Fvt5ptvRlVVVYvter0eN998c2drIiKiX7ydkoN/HW2629mmP4xud9hti0APZ9wRE4ikyQMR4uVi8/aJiOytQ4E3NTUVRmPLuWMNDQ344YcfOl0UERE1rYf75t6zAICXZw7D9ZE+dq6IiKh3ateUhuPHj1t/zszMhFartT43m83YvXs3goKCbFcdEVEfdST/EpZ8egwA8ND4cNwT39/OFRER9V7tCryxsbGQSCSQSCStTl1wdnbGO++8Y7PiiIj6oixtNR78x88wNlqQONgPy6cNtndJRES9WrumNOTl5SE3NxdCCBw6dAh5eXnWR1FREfR6Pe6///52F7FhwwaEhYVBpVIhISEBhw4duuK+mzdvxvjx4+Hp6QlPT08kJia22F8IgZUrVyIgIADOzs5ITExEdnZ2u+siIupu2aXV+P3mA6isNSImWIO37omFTHrlm0UQEdG1tSvwhoaGIiwsDBaLBXFxcQgNDbU+AgICIJPJ2l3A9u3bkZSUhFWrVuHIkSOIiYnBlClTcPHixVb3T01Nxdy5c7Fv3z6kpaUhJCQEkydPRlFRkXWfNWvW4O2338amTZtw8OBBuLq6YsqUKWhoaGh3fURE3SXnYg3mbj6IilojhgWp8cH9CXBVcjEdIqLOkgghRHsPSk5Ohp+fX4vR3Pfeew9lZWVYtmxZm9tKSEjAmDFjsH79egCAxWJBSEgInnzySSxfvvyax5vNZnh6emL9+vWYP38+hBAIDAzE4sWLsWTJEgCATqeDn58ftmzZgnvuueeaber1emg0Guh0OqjVXHydiLpeXnkt5vw1DRerDRgcoMbHDyXAw0Vh77KIiHqs9uS1Dq3S8Ne//hWDBg1qsX3o0KHYtGlTm9sxGo1IT09HYmLirwVJpUhMTERaWlqb2qirq4PJZIKXlxeApmkXWq22WZsajQYJCQlXbNNgMECv1zd7EBF1lwsVtZj7twO4WG3AIH93fPQgwy4RkS116G9lWq0WAQEBLbb7+vqipKSkze2Ul5fDbDbDz8+v2XY/Pz+cOXOmTW0sW7YMgYGB1oB7eeWI1tr87aoSv5WcnIwXXnihzXUTEbVXg8mMLzOKUFHbcknHrQfzodU3IKqfG/75YAK8XBl2iYhsqUOBNyQkBPv370d4eHiz7fv370dgYKBNCmuL1atXY9u2bUhNTYVKpepwOytWrEBSUpL1uV6vR0hIiC1KJCJCblkNntiagdMlV/7rUYSvKz56KAE+bspurIyIqG/oUOB96KGHsGjRIphMJuvyZCkpKVi6dCkWL17c5nZ8fHwgk8lQWlrabHtpaSn8/f2veuzatWuxevVq7N27FyNGjLBuv3xcaWlps1Ho0tJSxMbGttqWUqmEUslfMkRkW0IIfH6kCCv/dRJ1RjO8XBVIHNwPEjRfdcHD1QkPjAtHP/eO/487ERFdWYcC79NPP42Kigo89thj1juuqVQqLFu2DCtWrGhzOwqFAqNHj0ZKSgpmzpwJoOmitZSUFDzxxBNXPG7NmjV45ZVXsGfPHsTFxTV7LTw8HP7+/khJSbEGXL1ej4MHD+LRRx9t34kSUa9nsQiknavA8GAN1CqnbnvfGkMjnvvyJL7IaFpB5voB3nhzTiz81Ay1RETdrUOBVyKR4LXXXsNzzz2H06dPw9nZGVFRUR0aJU1KSsKCBQsQFxeH+Ph4rFu3DrW1tVi4cCEAYP78+QgKCkJycjIA4LXXXsPKlSuxdetWhIWFWeflurm5wc3NDRKJBIsWLcLLL7+MqKgohIeH47nnnkNgYKA1VBNR37Hxu1y8vicLARoV1t4VgxtsdHtei0Ugv7IO+ZV1MAsBCEBAwGIB6k1mvPFNFs5X1EEmleCpxCg8elMk19MlIrKTTi3wqNVqUVlZiQkTJkCpVEIIAYmkfV/oc+bMQVlZGVauXAmtVovY2Fjs3r3betFZfn4+pNJfF5PYuHEjjEYj7rzzzmbtrFq1Cs8//zwAYOnSpaitrcXDDz+MqqoqjBs3Drt37+7UPF8i6n0sFoGPD+UDAEp0DZj3/w5i4Q1hWDZ1EFRO7Vs3/Hx5LdIvXMLJYh1OFeuRWaxHjaHxqscEalR4e+5IxIV5dfgciIio8zq0Dm9FRQXuvvtu7Nu3DxKJBNnZ2YiIiMD9998PT09PvPHGG11Ra7fhOrxEjiEttwJzNx+Au1KO6bGB2HqwKfxG9nPDm3fHYniwpk3tfH2iBE98nAGzpfnXpUIuRbi3K5zkEkgggUSCptm5EgmGBLhj2dRBXF6MiKiLtCevdWiE96mnnoKTkxPy8/MxePCv93ifM2cOkpKSen3gJSLH8Fl6IQDg9phAvPq74bhliB+WfnYcORdr8Lt392NRYhQenxh51b9M1RvNePE/mTBbBIYFqTEmzAtDAzUYFqTGAF83OMk6tJw5ERF1ow4F3m+++QZ79uxBcHBws+1RUVG4cOGCTQojIuqMWkMjvj7ZtC74naObvqsmDuyHbxZNwDNfnsCuE1qs/eYsXBRy3D8u/IrtbP7hHEp0DQjycMZnj1zf7qkQRERkfx0amqitrYWLi0uL7ZWVlVzei4h6hF0nSlBnNCPCxxWj+ntYt3u6KrDh96OwYlrT3SJf230GORdrWm2jVN+Ajam5AIBl09o/75eIiHqGDgXe8ePH44MPPrA+l0gksFgsWLNmDSZOnGiz4oiIOurydIbZo4NbTFmQSCR4eEIExkf5wNBoweJPj6HRbGnRxto9Wag3mTGqvwemj2h5d0kiIuodOjSlYc2aNZg0aRJ+/vlnGI1GLF26FKdOnUJlZSX2799v6xqJiNolv6IOB/MqIZEAs0YFtbqPRCLBmjtHYPKb3+NYQRU2pubiyUlR1tdPFunw2ZGm0Pzs7UPavQINERH1HB0a4R02bBjOnj2LcePGYcaMGaitrcWsWbOQkZGBAQMG2LpGIqJ2uRxUx0X6IEDjfMX9AjTOeHHGUADAWynZOFmkA9B0h7SXd2ZCCOCOmECM6u/Z9UUTEVGXafMI76xZs7Blyxao1Wp88MEHmDNnDp555pmurI2IqN0sFoHPf5nOcPlitauZGRuEPSdLsfuUFos/OYavnrwB32WV4cC5SijlUiz7Za4vERH1Xm0e4f3Pf/6D2tpaAMDChQuh0+m6rCgioo46kFeBoqp6uKvkmDLU/5r7SyQSvPK7YfB2VSCrtBprdmch+eszAICHxkcgyOPKI8RERNQ7tHmEd9CgQVixYgUmTpwIIQQ++eSTKy7yO3/+fJsVSETUHpcvVpseE9jmVRW83ZR4ddZw/PHDdPz9xzwAgK+7Eo/cxClaRESOoM2Bd+PGjVi8eDF27twJiUSCZ599ttWLOCQSCQMvEXWp/Io6FOvqMTrUs9mNH2oMjfj6hBZA26Yz/NaUof6YNSoIO44UAQCWTI6Gm7JTd18nIqIeos3f5jfccAMOHDgAAJBKpTh79iz69evXZYUREf2vi9UNWLc3G9sPF8BsEfB2VWDmyCDcOToYgwPU2HWiBPUmMyJ8XTEyxKPd7a+aPhS5F2vg5arAnaNDbH8CRERkFx26aO3999+Hu7t7V9ZFRGRVa2jE374/h80/nEOd0QwAcFfJUVFrxN9/zMPff8zDsCA16gxNr93Zytq7baFxdsK/nhhn09qJiMj+JEII0ZYdFQoFLly4gICAAMhkMpSUlDjsCK9er4dGo4FOp7viPGUi6rwGkxk5F2uQpa1GTlkNhGgKspcfbkonFFfV451vc1BeYwAAxIZ44M+3DsbI/h74/mwZPksvxN7TpTCZm77KpBLgp+WT4K9R2fPUiIioi7Unr/GiNSLqckIIFOsacLpYj8wSPc5o9cjSVuN8RR3Mljb9PzfCvF2wdOogTBvmbx29nTTYD5MG+6Gy1oivjhbh65NaXBfhzbBLRETNtHmE96effkJSUhJyc3NRWVkJd3f3K160VllZafNCuxNHeIk67lKtETllNci5WIPs0hpkluhwuqQaunpTq/trnJ0w0N8dA/3coZBLUd1gQo2hEdUNTQ+LEJg9Khhz4/tDIe/QvXKIiMgBtSevtTnw/pZUKoVWq+WUBqI+ythoQcGlOpwvr0VeeS3Oldci52INci/WoKLW2OoxcqkEkf3cMCRAjUEB7hjor8Ygf3f0c1fytr1ERNRuXTKl4bfy8vKQlZWFpKQknDt3Dp9++imCgoLw4YcfIjw8HOPG8aIPIkcjhMCru07jv5mlKLhUf9WpCEEezhjQzw2Rvm4YHOCOwQFqRPm5QSlv27q4REREttShwPvzzz/j3nvvxbx583DkyBEYDE0Xk+h0Orz66qvYtWuXTYskIvtLO1eBzT/kWZ87O8kQ7uOKcF9XhHu7YkA/V0T6uiPC1xWuXL+WiIh6kA79Vnr55ZexadMmzJ8/H9u2bbNuv+GGG/Dyyy/brDgi6jk27MsBAMweFYynpwyEn5pTEYiIqHfoUODNysrChAkTWmzXaDSoqqrqbE1E1MNk5F/C/pwKyKUSPHVLFFdBICKiXqVDlzz7+/sjJyenxfYff/wRERERnS6KiHqWd1NzAQAzRwYh2NPFztUQERG1T4cC70MPPYT/+7//w8GDByGRSFBcXIyPPvoIS5YswaOPPmrrGonIjs5o9fhvZikkEuCRGwfYuxwiIqJ269CUhuXLl8NisWDSpEmoq6vDhAkToFQqsWTJEjz55JO2rpGI7GjjL6O704b5I7Kfm52rISIiar8OrcN7mdFoRE5ODmpqajBkyBC4uTnGL0Ouw0vU5EJFLSauTYVFAP95chyGBWnsXRIRERGAbliH9zKFQoEhQ4Z0pgki6sE2fXcOFgHcNNCXYZeIiHot3qeTiFql1TXg8/RCAMDjEyPtXA0REVHHMfASUas2/3AORrMF8WFeGBPmZe9yiIiIOoyBl4haqKw1YuvBfADA4zdzdJeIiHo33v+TqI/R1ZtwqliHU0V6nCzWIUtbDaPZ0myfWkMj6k1mDAtSY0KUj50qJSIisg0GXiIHZrEInL1YjcN5lTh0/hKOFVQhv7Kuzcf/36Ro3j6YiIh6PQZeoh7kor4BRwuqUKpvQKneAK2+AaX6BpRVG9DfywWzRgXh5kF+UMhbn41ksQhklujxU245DuVV4vD5S9DVm1rsF+zpjGGBGgwLUmNwgBpuypZfBR4uCgz0d7f5ORIREXU3Bl6idvog7Tw2/3AOq24fisQhfp1uz2IR+DGnHB8dvIC9py/CbGl9aewz2mp8k1kKDxcn3BETiNmjgjEiWINLdSb8kF2G786W4fuz5SivMTQ7zkUhw+hQT4wJ88Ko/p4YFqSGh4ui03UTERH1Fp268YSj4o0n6EpOFeswY/1+NFoEFDIp3l84BjdEdmyOa0WNAZ+mF2Lrwfxm0wwGB6gR4ukMf40KfmoV/NUqeLkpcOBcBb7MKEKp/tdA66dW4mK1Ab/9V+yikGFshDeui/BGfLgXhgSq4STj9alERORY2pPXGHhbwcBLrTGZLbhj/X6cLtHDXSlHtaERLgoZ/vlgAkb192xzOw0mM95Kycbff8izXizmrpJj9qhg/D6hP6L9rjyNwGwR2J9Tjs+PFGLPKS0aTE3HD/J3x40DfXFjtC/iQr2uOOWBiIjIUXTbndaI+pJ39+XidIkeni5O2Pmn8Vj2+XH8kF2O+947hO1/HIvBAdf+n6P0C5V4+rPjOFdWCwCICdZgXkIobo8JgIvi2v8cZVIJJkT7YkK0L6obTDhaUIWofu7w16g6fX5ERESOiiO8reAIL/2v0yV63LH+R5jMAm/dE4sZsUGoMzbi3r8fQvqFS/BxU+LTR8Yi3Me11ePrjI1Yu+cs3v8pD0IAvu5KvDxzGKYM9e/mMyEiInIM7clr/Lsn0TWYzBY8/dkxmMwCk4f44Y6YQACAi0KO9+4bg8EBapTXGPCH/3cQxVX1sFgEqhtM0OoakHOxBv/NLMXUdT/gvf1NYffO0cHY+9SNDLtERETdhFMaiK7hr9/l4mSRHhpnJ7z8u2HN1qXVODvhwwficfemNJwrr8WENfvQeIVVFgI1Krw6azhuGtivu0onIiIicISX6KqytNV4KyUbAPD8HUPQz73lXFkfNyU+fDABod4uzcKuVNJ0MVqgRoUFY0Ox56kJDLtERER2YPcR3g0bNuD111+HVqtFTEwM3nnnHcTHx7e676lTp7By5Uqkp6fjwoULePPNN7Fo0aJm+zz//PN44YUXmm0bOHAgzpw501WnQA7gZJEOJ4p0Lbb/88AFmMwCkwb1w8zYoCseH+ThjJSkG1Fc1QAXpQxuSjmUcinvUkZERNQD2DXwbt++HUlJSdi0aRMSEhKwbt06TJkyBVlZWejXr+VIWF1dHSIiInDXXXfhqaeeumK7Q4cOxd69e63P5XK753rqwaobTLhrUxrqTeZWX1er5Hh11vBrhle5TIr+3i5dUSIRERF1gl2T4F/+8hc89NBDWLhwIQBg06ZN2LlzJ9577z0sX768xf5jxozBmDFjAKDV1y+Ty+Xw9+cFQdQ2GflVqDeZ4a6SIyHcu9lrMikwLyEUfmou+0VERNRb2S3wGo1GpKenY8WKFdZtUqkUiYmJSEtL61Tb2dnZCAwMhEqlwtixY5GcnIz+/ftfcX+DwQCD4de7V+n1+k69P/UuP1+4BABIHOyHN+fE2rcYIiIisjm7XbRWXl4Os9kMPz+/Ztv9/Pyg1Wo73G5CQgK2bNmC3bt3Y+PGjcjLy8P48eNRXV19xWOSk5Oh0Wisj5CQkA6/P/U+6RcqAQCjQ9t+tzQiIiLqPRxulYZp06bhrrvuwogRIzBlyhTs2rULVVVV+OSTT654zIoVK6DT6ayPgoKCbqyY7KnRbEFGfhUAIC6MgZeIiMgR2W1Kg4+PD2QyGUpLS5ttLy0tten8Ww8PD0RHRyMnJ+eK+yiVSiiVSpu9J/UeZ7TVqDM2zd+N7udu73KIiIioC9hthFehUGD06NFISUmxbrNYLEhJScHYsWNt9j41NTXIzc1FQECAzdokx3H4fNN0hlH9PSGVcgkxIiIiR2TXVRqSkpKwYMECxMXFIT4+HuvWrUNtba111Yb58+cjKCgIycnJAJoudMvMzLT+XFRUhKNHj8LNzQ2RkZEAgCVLlmD69OkIDQ1FcXExVq1aBZlMhrlz59rnJKlHu3zBWhzn7xIRETksuwbeOXPmoKysDCtXroRWq0VsbCx2795tvZAtPz8fUumvg9DFxcUYOXKk9fnatWuxdu1a3HjjjUhNTQUAFBYWYu7cuaioqICvry/GjRuHAwcOwNfXt1vPjXo+IQTSzzcF3tGcv0tEROSwJEIIce3d+ha9Xg+NRgOdTge1Wm3vcqiLFF6qw7jX9kEmleDE85PhouANSoiIiHqL9uQ1h1ulgait0n+ZzjA0UM2wS0RE5MAYeKnP+vnydAbO3yUiInJoDLzUZ12+YG1MmJedKyEiIqKuxMBLfVJ1gwlZ2qZbSHOFBiIiIsfGwEt9UkZ+FSwCCPFyRj+1yt7lEBERURdi4KU+6df1dzmdgYiIyNEx8FKflH6h6Q5rvGCNiIjI8THwUp/TaLYgI78KABDHG04QERE5PAZe6nNOl1SjzmiGu0qO6H7u9i6HiIiIuhgDL/U5P/8ynWFUf09IpRI7V0NERERdjYGX+pxfL1jjdAYiIqK+gIGX+hQhBNIv32GN83eJiIj6BLm9CyBqqxpDI7K01VDKpXBRyOCskMHFSQ5nhQwSCdBoFjCaLWg0W2AyC1iEgL9a1WzaQlFVPbT6BsikEsSGeNjvZIiIiKjbMPCSzQkhIJHYbm6s2SKw7XA+1u7JwqU6U7uOVavkiA/3xnURXkgI90b2xWoAwNBANVwU/PgTERH1BfyNTzb16c8FWPr5ccSHeeHesaGYPMQfCnnHZ84cyqvE81+dQmZJ022AvV0VkEklqDeZUW80o9EiWj1O/suorr6hEXtPl2Lv6dJmr3P9XSIior6DgZdsaueJEggBHMyrxMG8Svi6KzF3TAjmJvRHgMa5ze0UV9Uj+esz+PexYgBNI7VJt0TjD9eFQi77NUAbGy2oN5oBAHKZBE4yKeRSCaRSCRrNFpws1uPguQoczKvE4bxKVBsaAQATon1teNZERETUk0mEEK0PkfVher0eGo0GOp0OarXa3uX0KmOTU1Cia8DvRgbhx5xylFUbAAAyqQQjgjWwCMBgMsPYaIGh0QJDoxmtDdJWN5hgMgtIJMDc+P5YfEs0vN2UnarNbBHILNaj2mDC9QN8OtUWERER2Vd78hpHeMlmdHUmlOgaAAAvzBgKZycZvjlVig8PnMeBc5XWu5u11ZgwT6yaPhTDgjQ2qU8mlWB4sG3aIiIiot6DgZdsJqu06YKwIA9nqFVOAIDbRgTgthEByLlYY11hQekkhVIus/4sa+UCN4Vciv5eLja9+I2IiIj6JgZespksbdOFZQP9W96uN7KfGyL7uXV3SURERES88QTZzmlt0whva4GXiIiIyF4YeMlmsn4JvIMYeImIiKgHYeAlmxBC4CxHeImIiKgHYuAlmyiqqke1oRFyqQQRPpyrS0RERD0HAy/ZxOXpDAN83Tp1ZzUiIiIiW2MyIZs4w+kMRERE1EMx8JJNZDHwEhERUQ/FwEs2wRUaiIiIqKdi4KVOMzZakFtWAwAYFHD1e1kTERERdTcGXuq0c+U1aLQIuKvkCNSo7F0OERERUTMMvNRp1vm7fu6QSCR2roaIiIioOQZe6jSu0EBEREQ9GQMvdRovWCMiIqKejIGXOu3XJcl4wRoRERH1PAy8Du6n3HIcOFfRZe3rG0woqqoH0DSHl4iIiKinYeB1YMcKqvD7zQdxz98O4OND+V3yHmd/Gd0N0KigcXHqkvcgIiIi6gwGXgclhMAL/z5lff7nL07gi4zCdrfTaLbgb9/nYv232bBYRIvXT/OCNSIiIurh5PYugLrGV8eKcSS/Ci4KGaYO9ceOjCIs+fQ4VHIZpg0PaFMbFTUGPLE1A2m/TIkI9nTBzJFBzfbJ0uoBMPASERFRz2X3Ed4NGzYgLCwMKpUKCQkJOHTo0BX3PXXqFGbPno2wsDBIJBKsW7eu0206ojpjI5J3nQEAPD4xEmvvisFdo4Nhtgj8aVsGvj1Tes02jhVUYfo7PyLtXAUuL627+uszqDM2NtuPKzQQERFRT2fXwLt9+3YkJSVh1apVOHLkCGJiYjBlyhRcvHix1f3r6uoQERGB1atXw9/f3yZtOqJNqbnQ6hsQ7OmMB8aFQyqVYPXsEZgeEwiTWeCRfx7B/pzyKx6//XA+7tqUhmJdAyJ8XPHvJ8Yh2NMZWn0DNn13zrqfEMK6Bu8grtBAREREPZRECNFyYmY3SUhIwJgxY7B+/XoAgMViQUhICJ588kksX778qseGhYVh0aJFWLRokc3avEyv10Oj0UCn00Gt7l1BrqCyDol/+Q6GRgs2/WEUpg77dfqCyWzB4x8dwTeZpXB2kmH+9aHwclFA7ewEjbMT1Con7DpZgq0Hmy5wu2WIH964O6Zp+4kSPPbRESjlUny75CYEeTijuKoe16/+FnKpBJkvToVCbvc/GBAREVEf0Z68Zrc5vEajEenp6VixYoV1m1QqRWJiItLS0npMm73N6q/PwNBowdgIb0wZ2nwU3EkmxTu/H4mHP0jHd2fL8NffjNb+lkQCLL4lGo/dFAmptGk+w7Rh/ogP98KhvEq89vUZvD13pHU6Q4SvK8MuERER9Vh2C7zl5eUwm83w8/Nrtt3Pzw9nzpzp1jYNBgMMBoP1uV6v79D729uBcxXYeaIEUgmwcvoQSC5Pvv0NpVyGv947GtsO5eN8RR30DSbo603Q1Zugr2+Ek1yCxZMHYuLAfs2Ok0gkWHn7EExf/yO+OlaMBdeH/uaWwr1rFJyIiIj6Fq7SACA5ORkvvPCCvcvoFLNF4MV/ZwIA5sb3x+CAK4dQlZMM990Q3u73GBakwd2jQ7D95wK88O9MhHm7AuAFa0RERNSz2e3v0D4+PpDJZCgtbb5iQGlp6RUvSOuqNlesWAGdTmd9FBQUdOj97aXBZMZbKdnILNFDrZIj6ZboLnuvJVMGwk0px/FCHb4+WQKAd1gjIiKins1ugVehUGD06NFISUmxbrNYLEhJScHYsWO7tU2lUgm1Wt3s0RtU1Rmx/ttsjHttH95OyQYALEqMhrebssve09ddiSdujgQAmMxN1ztyDV4iIiLqyew6pSEpKQkLFixAXFwc4uPjsW7dOtTW1mLhwoUAgPnz5yMoKAjJyckAmi5Ky8zMtP5cVFSEo0ePws3NDZGRkW1q0xEUVNbh7z/mYfvhAtSbzACAQI0KD02IwIKxYV3+/gtvCMPWg/nIr6yDm1KOYE/nLn9PIiIioo6ya+CdM2cOysrKsHLlSmi1WsTGxmL37t3Wi87y8/Mhlf46CF1cXIyRI0dan69duxZr167FjTfeiNTU1Da12dvtOaXFYx8dgfmX2/wOCVDj4QkRuG1EAJxk3TNgr5TL8NztQ/DQBz8jIdyr1YvjiIiIiHoKu67D21P11HV4S/UNmPzm99DVm3BdhBeemBiFGyK97RY4z5ZWw0+tgsbZyS7vT0RERH1Xr1iHl9pHCIGnPzsOXb0Jw4M0+PCBhG4b0b2SaF6sRkRERL0A7xbQS3x0MB/fny2DQi7Fm3Ni7B52iYiIiHoLpqZeIK+8Fq/sPA0AWDZ1ECL7cWSViIiIqK0YeHu4RrMFSZ8cRb3JjOsHeGPh9WH2LomIiIioV2Hg7eE2fZeLjPwquCvleP2uGEilXBGBiIiIqD0YeHuwk0U6rNvbdEOJ5+8YiiAPrndLRERE1F5cpaGHMDSakV9Rh9yyWuSV1yKvvAbfny1Ho0Vg6lB/zBoVZO8SiYiIiHolBt4e4PZ3fkBmsR6WVlZE9lMr8crvhvHmDkREREQdxMDbA8ikUlgE4K6UI8LXFeE+rgj3cUO4rysmRPnAw0Vh7xKJiIiIei0G3h7gL3fHQK1ygo+bgiO5RERERDbGwNsDDPB1s3cJRERERA6LqzQQERERkUNj4CUiIiIih8bAS0REREQOjYGXiIiIiBwaAy8REREROTQGXiIiIiJyaAy8REREROTQuA5vK4RousevXq+3cyVERERE1JrLOe1ybrsaBt5WVFdXAwBCQkLsXAkRERERXU11dTU0Gs1V95GItsTiPsZisaC4uBju7u42udWvXq9HSEgICgoKoFarbVAhXca+7Trs267Dvu067Nuuw77tOuzbjhFCoLq6GoGBgZBKrz5LlyO8rZBKpQgODrZ5u2q1mh/kLsK+7Trs267Dvu067Nuuw77tOuzb9rvWyO5lvGiNiIiIiBwaAy8REREROTQG3m6gVCqxatUqKJVKe5ficNi3XYd923XYt12Hfdt12Lddh33b9XjRGhERERE5NI7wEhEREZFDY+AlIiIiIofGwEtEREREDo2Bl4iIiIgcGgNvB23YsAFhYWFQqVRISEjAoUOHrrr/p59+ikGDBkGlUmH48OHYtWtXs9fvu+8+SCSSZo+pU6d25Sn0WO3p21OnTmH27NkICwuDRCLBunXrOt2mI7N13z7//PMtPreDBg3qwjPoudrTt5s3b8b48ePh6ekJT09PJCYmtthfCIGVK1ciICAAzs7OSExMRHZ2dlefRo9j637ld+2v2tO3O3bsQFxcHDw8PODq6orY2Fh8+OGHzfbhZ/ZXtu5bfm5tQFC7bdu2TSgUCvHee++JU6dOiYceekh4eHiI0tLSVvffv3+/kMlkYs2aNSIzM1M8++yzwsnJSZw4ccK6z4IFC8TUqVNFSUmJ9VFZWdldp9RjtLdvDx06JJYsWSI+/vhj4e/vL958881Ot+mouqJvV61aJYYOHdrsc1tWVtbFZ9LztLdvf//734sNGzaIjIwMcfr0aXHfffcJjUYjCgsLrfusXr1aaDQa8eWXX4pjx46JO+64Q4SHh4v6+vruOi2764p+5Xdtk/b27b59+8SOHTtEZmamyMnJEevWrRMymUzs3r3bug8/s026om/5ue08Bt4OiI+PF48//rj1udlsFoGBgSI5ObnV/e+++25x2223NduWkJAg/vjHP1qfL1iwQMyYMaNL6u1N2tu3vxUaGtpqKOtMm46kK/p21apVIiYmxoZV9k6d/Yw1NjYKd3d38Y9//EMIIYTFYhH+/v7i9ddft+5TVVUllEql+Pjjj21bfA9m634Vgt+1l9nie3HkyJHi2WefFULwM/tbtu5bIfi5tQVOaWgno9GI9PR0JCYmWrdJpVIkJiYiLS2t1WPS0tKa7Q8AU6ZMabF/amoq+vXrh4EDB+LRRx9FRUWF7U+gB+tI39qjzd6oK/shOzsbgYGBiIiIwLx585Cfn9/ZcnsVW/RtXV0dTCYTvLy8AAB5eXnQarXN2tRoNEhISOgzn9uu6NfL+F3bub4VQiAlJQVZWVmYMGECAH5mL+uKvr2sr39uO0tu7wJ6m/LycpjNZvj5+TXb7ufnhzNnzrR6jFarbXV/rVZrfT516lTMmjUL4eHhyM3NxZ///GdMmzYNaWlpkMlktj+RHqgjfWuPNnujruqHhIQEbNmyBQMHDkRJSQleeOEFjB8/HidPnoS7u3tny+4VbNG3y5YtQ2BgoPWX5OXvhmt9bziyruhXgN+1QMf7VqfTISgoCAaDATKZDO+++y5uueUWAPzMXtYVfQvwc2sLDLw9xD333GP9efjw4RgxYgQGDBiA1NRUTJo0yY6VEV3ZtGnTrD+PGDECCQkJCA0NxSeffIIHHnjAjpX1HqtXr8a2bduQmpoKlUpl73IcxpX6ld+1Hefu7o6jR4+ipqYGKSkpSEpKQkREBG666SZ7l9brXatv+bntPE5paCcfHx/IZDKUlpY2215aWgp/f/9Wj/H392/X/gAQEREBHx8f5OTkdL7oXqIjfWuPNnuj7uoHDw8PREdH83OLtvXt2rVrsXr1anzzzTcYMWKEdfvl4/ry57Yr+rU1/K791bX6ViqVIjIyErGxsVi8eDHuvPNOJCcnA+Bn9rKu6NvW9MXPbWcx8LaTQqHA6NGjkZKSYt1msViQkpKCsWPHtnrM2LFjm+0PAP/973+vuD8AFBYWoqKiAgEBAbYpvBfoSN/ao83eqLv6oaamBrm5ufzctqFv16xZg5deegm7d+9GXFxcs9fCw8Ph7+/frE29Xo+DBw/2mc9tV/Rra/hd26Qj3wcWiwUGgwEAP7OXdUXftqYvfm47zd5XzfVG27ZtE0qlUmzZskVkZmaKhx9+WHh4eAitViuEEOLee+8Vy5cvt+6/f/9+IZfLxdq1a8Xp06fFqlWrmi1LVl1dLZYsWSLS0tJEXl6e2Lt3rxg1apSIiooSDQ0NdjlHe2lv3xoMBpGRkSEyMjJEQECAWLJkicjIyBDZ2dltbrOv6Iq+Xbx4sUhNTRV5eXli//79IjExUfj4+IiLFy92+/nZU3v7dvXq1UKhUIjPPvus2TJD1dXVzfbx8PAQ//rXv8Tx48fFjBkz+twST7buV37X/qq9ffvqq6+Kb775RuTm5orMzEyxdu1aIZfLxebNm6378DPbxNZ9y8+tbTDwdtA777wj+vfvLxQKhYiPjxcHDhywvnbjjTeKBQsWNNv/k08+EdHR0UKhUIihQ4eKnTt3Wl+rq6sTkydPFr6+vsLJyUmEhoaKhx56qM8Fssva07d5eXkCQIvHjTfe2OY2+xJb9+2cOXNEQECAUCgUIigoSMyZM0fk5OR04xn1HO3p29DQ0Fb7dtWqVdZ9LBaLeO6554Sfn59QKpVi0qRJIisrqxvPqGewZb/yu7a59vTtM888IyIjI4VKpRKenp5i7NixYtu2bc3a42f2V7bsW35ubUMihBDdO6ZMRERERNR9OIeXiIiIiBwaAy8REREROTQGXiIiIiJyaAy8REREROTQGHiJiIiIyKEx8BIRERGRQ2PgJSIiIiKHxsBLRNQDnD9/HhKJBEePHr3iPqmpqZBIJKiqquqWmp5//nnExsZ2y3sREXUlub0LICIiICQkBCUlJfDx8bF3KUREDoeBl4jIzoxGIxQKBfz9/e1dChGRQ+KUBiIiG6uursa8efPg6uqKgIAAvPnmm7jpppuwaNEiAEBYWBheeuklzJ8/H2q1Gg8//HCrUxp27dqF6OhoODs7Y+LEiTh//nyb3l+v18PZ2Rlff/11s+1ffPEF3N3dUVdXBwBYtmwZoqOj4eLigoiICDz33HMwmUxXbPe353DZzJkzcd9991mfGwwGLFmyBEFBQXB1dUVCQgJSU1Otr1+4cAHTp0+Hp6cnXF1dMXToUOzatatN50VE1FEMvERENpaUlIT9+/fjq6++wn//+1/88MMPOHLkSLN91q5di5iYGGRkZOC5555r0UZBQQFmzZqF6dOn4+jRo3jwwQexfPnyNr2/Wq3G7bffjq1btzbb/tFHH2HmzJlwcXEBALi7u2PLli3IzMzEW2+9hc2bN+PNN9/s4Fk3eeKJJ5CWloZt27bh+PHjuOuuuzB16lRkZ2cDAB5//HEYDAZ8//33OHHiBF577TW4ubl16j2JiK6FUxqIiGyouroa//jHP7B161ZMmjQJAPD+++8jMDCw2X4333wzFi9ebH3+v6O3GzduxIABA/DGG28AAAYOHGgNiG0xb9483Hvvvairq4OLiwv0ej127tyJL774wrrPs88+a/05LCwMS5YswbZt27B06dJ2nfNl+fn5eP/995Gfn2893yVLlmD37t14//338eqrryI/Px+zZ8/G8OHDAQAREREdei8iovZg4CUisqFz587BZDIhPj7euk2j0WDgwIHN9ouLi7tqO6dPn0ZCQkKzbWPHjm1zHbfeeiucnJzw1Vdf4Z577sHnn38OtVqNxMRE6z7bt2/H22+/jdzcXNTU1KCxsRFqtbrN7/G/Tpw4AbPZjOjo6GbbDQYDvL29AQB/+tOf8Oijj+Kbb75BYmIiZs+ejREjRnT4PYmI2oJTGoiI7MDV1bVL21coFLjzzjut0xq2bt2KOXPmQC5vGudIS0vDvHnzcOutt+I///kPMjIy8Mwzz8BoNF6xTalUCiFEs22/nfNbU1MDmUyG9PR0HD161Po4ffo03nrrLQDAgw8+iHPnzuHee+/FiRMnEBcXh3feecfWp09E1AwDLxGRDUVERMDJyQmHDx+2btPpdDh79my72hk8eDAOHTrUbNuBAwfa1ca8efOwe/dunDp1Ct9++y3mzZtnfe2nn35CaGgonnnmGcTFxSEqKgoXLly4anu+vr4oKSmxPjebzTh58qT1+ciRI2E2m3Hx4kVERkY2e/x2BYqQkBA88sgj2LFjBxYvXozNmze367yIiNqLgZeIyIbc3d2xYMECPP3009i3bx9OnTqFBx54AFKpFBKJpM3tPPLII8jOzsbTTz+NrKwsbN26FVu2bGlXLRMmTIC/vz/mzZuH8PDwZlMkoqKikJ+fj23btiE3Nxdvv/12s/m9rbn55puxc+dO7Ny5E2fOnMGjjz7a7CYY0dHRmDdvHubPn48dO3YgLy8Phw4dQnJyMnbu3AkAWLRoEfbs2YO8vDwcOXIE+/btw+DBg9t1XkRE7cXAS0RkY3/5y18wduxY3H777UhMTMQNN9yAwYMHQ6VStbmN/v374/PPP8eXX36JmJgYbNq0Ca+++mq76pBIJJg7dy6OHTvWbHQXAO644w489dRTeOKJJxAbG4uffvqp1dUifuv+++/HggULMH/+fNx4442IiIjAxIkTm+3z/vvvY/78+Vi8eDEGDhyImTNn4vDhw+jfvz+AplHhxx9/HIMHD8bUqVMRHR2Nd999t13nRUTUXhLxvxOyiIjIpmpraxEUFIQ33ngDDzzwgL3LISLqc7hKAxGRjWVkZODMmTOIj4+HTqfDiy++CACYMWOGnSsjIuqbOKWBiKgLXL6xRGJiImpra/HDDz/Ax8fHZu1PmzYNbm5urT7aO/WBiMjRcUoDEVEvVFRUhPr6+lZf8/LygpeXVzdXRETUczHwEhEREZFD45QGIiIiInJoDLxERERE5NAYeImIiIjIoTHwEhEREZFDY+AlIiIiIofGwEtEREREDo2Bl4iIiIgcGgMvERERETm0/w/CcEa/ckvw8QAAAABJRU5ErkJggg==",
      "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('04_PLOT_pdp_naics_meanenc.png'),\n",
    "               bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5488c384-feb9-45b4-8d95-e2ea4954719b",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "83c4bf72-5710-489a-888f-bd2b7590e93b",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e2b6ef2b-9fe4-4595-9ba9-7fc205a864f6",
   "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
}