23_nn_menc.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8816b366-5be4-4748-b769-4dc6f7d1735f",
   "metadata": {},
   "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": "a945de0c-f042-46b2-9d06-ab064af6c05b",
   "metadata": {},
   "source": [
    "# Neural Network - With Target Encoded Feature\n",
    "Neural network plus the mean encoded feature"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d47a7ec9-7d7b-47df-a902-b2f1e54f1135",
   "metadata": {},
   "source": [
    "*This script takes about 2 hours on my MacBook Air*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "fb42f9b4-99a8-4753-aacf-8722469be4c8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "bf51170a-4487-4d10-b63d-c88f41b529f9",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "import importlib, pickle\n",
    "from sklearn import model_selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "09dd973b-60a0-434b-9193-8c5fcec27dd5",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "import re\n",
    "\n",
    "from tensorflow.keras import layers, optimizers, losses, metrics, Model\n",
    "from sklearn import preprocessing, feature_extraction, model_selection\n",
    "from IPython.display import display, HTML\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "22b466e7-a419-45aa-83f7-5dafc7a3998b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Dropout\n",
    "import tensorflow.keras.metrics as km"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9f1e4f2e-0e45-4e28-8e9c-212e0bb51199",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sba_gnn.sba_gnn import sg_plot "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3cbc00d2-bc83-42d9-83eb-65fcf306dccd",
   "metadata": {},
   "source": [
    "## Input Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5f0183dc-7ee1-49eb-8723-a9f4cea14b88",
   "metadata": {},
   "outputs": [],
   "source": [
    "business_data = pd.read_parquet(Path(setup.temp_path).joinpath('20_DATA_combined_scaled_all.parquet'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "555f4031-e66d-4476-a37e-50e2d382ecf4",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(Path(setup.temp_path).joinpath('20_DATA_features.pkl'), 'rb') as fin:\n",
    "    imputer_features = pickle.load(fin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3a916e26-2419-4bd0-8fd4-58a959dc4bac",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['NoEmp', 'CreateJob', 'LowDoc', 'DisbursementGross', 'new_business', 'urban_flag', 'franchise_flag', 'missingindicator_LowDoc', 'missingindicator_new_business', 'missingindicator_urban_flag', 'menc_NAICS']\n"
     ]
    }
   ],
   "source": [
    "features = [f for f in imputer_features if 'NAICS' not in f]\n",
    "features = features + ['menc_NAICS']\n",
    "print(features)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d94006a0-f010-4874-a7f1-0fd80c1e9a35",
   "metadata": {},
   "source": [
    "##### Datasets for train, validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "a02a909d-834a-46a8-9cdd-ccfa36b07d69",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = business_data[['dset', 'LoanNr_ChkDgt'] + features].set_index('LoanNr_ChkDgt').sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "42dc5c92-f5a2-457c-be3a-60cb0f6c2ef8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training X: (436120, 11), y:(436120,)\n"
     ]
    }
   ],
   "source": [
    "X_train = X[X['dset'] == 'train'].drop(columns='dset')\n",
    "y_train = business_data[business_data['dset'] == 'train'].set_index('LoanNr_ChkDgt').sort_index()['target']\n",
    "print(f'training X: {X_train.shape}, y:{y_train.shape}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "038b0e4f-a84a-4c8b-863e-4c594c1c9f32",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "val X: (93454, 11), y:(93454,)\n"
     ]
    }
   ],
   "source": [
    "X_val = X[X['dset'] == 'val'].drop(columns='dset')\n",
    "y_val = business_data[business_data['dset'] == 'val'].set_index('LoanNr_ChkDgt').sort_index()['target']\n",
    "print(f'val X: {X_val.shape}, y:{y_val.shape}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "bcb86909-e8a7-4f22-b089-6ca61dd86c73",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.2046546821975603\n"
     ]
    }
   ],
   "source": [
    "base_thresh = y_train.mean()\n",
    "print(base_thresh)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5c68623-4599-4472-a13a-9400d101633e",
   "metadata": {},
   "source": [
    "## Function to create model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "9f4deffd-2b85-43d0-9cd1-1d9cf707b9e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_model(n_feat = len(features), \n",
    "                 hidden_size = setup.nn_layer_sizes,\n",
    "                 activation='tanh', lr=setup.nn_learning_rate,\n",
    "                 opt_func = setup.nn_optimizer, dropout = setup.nn_dropout):\n",
    "    \n",
    "    n_layers = len(hidden_size)\n",
    "    model = Sequential()\n",
    "    model.add(Dense(hidden_size[0], input_shape=(n_feat,), activation=activation))\n",
    "    for i in range(1, n_layers):\n",
    "        model.add(Dropout(dropout))\n",
    "        model.add(Dense(hidden_size[i], input_shape=(hidden_size[i-1],), activation=activation))\n",
    "    model.add(Dropout(dropout))\n",
    "    model.add(Dense(1, activation='sigmoid'))\n",
    "    # Compile model\n",
    "    optimizer = opt_func(learning_rate=lr)\n",
    "    model.compile(loss='binary_crossentropy', optimizer=optimizer, \n",
    "                  metrics=[km.AUC(curve='PR'), km.AUC(curve='ROC')])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3db4ea1c-77b2-4721-b106-1e66fc502139",
   "metadata": {},
   "source": [
    "## Create, fit model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4d48ddfa-580c-4321-9917-0bd176b67d87",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-03-28 22:16:45.540178: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M1\n",
      "2024-03-28 22:16:45.540208: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 16.00 GB\n",
      "2024-03-28 22:16:45.540214: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 5.33 GB\n",
      "2024-03-28 22:16:45.540675: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:303] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.\n",
      "2024-03-28 22:16:45.541081: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:269] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)\n"
     ]
    }
   ],
   "source": [
    "this_model = create_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "bb765db3-1646-44f2-84ec-f8f4a7b236ce",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 128)               1536      \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 128)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 64)                8256      \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 64)                0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 1)                 65        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 9857 (38.50 KB)\n",
      "Trainable params: 9857 (38.50 KB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "this_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "31772016-3643-468b-8d65-c456168cad6e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-03-28 22:16:46.292185: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13629/13629 [==============================] - ETA: 0s - loss: 0.4664 - auc: 0.3364 - auc_1: 0.6970"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-03-28 22:20:29.347809: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13629/13629 [==============================] - 252s 18ms/step - loss: 0.4664 - auc: 0.3364 - auc_1: 0.6970 - val_loss: 0.4629 - val_auc: 0.3615 - val_auc_1: 0.7106\n",
      "Epoch 2/20\n",
      "13629/13629 [==============================] - 248s 18ms/step - loss: 0.4584 - auc: 0.3586 - auc_1: 0.7140 - val_loss: 0.4604 - val_auc: 0.3695 - val_auc_1: 0.7160\n",
      "Epoch 3/20\n",
      "13629/13629 [==============================] - 250s 18ms/step - loss: 0.4569 - auc: 0.3632 - auc_1: 0.7170 - val_loss: 0.4608 - val_auc: 0.3723 - val_auc_1: 0.7187\n",
      "Epoch 4/20\n",
      "13629/13629 [==============================] - 178s 13ms/step - loss: 0.4563 - auc: 0.3656 - auc_1: 0.7182 - val_loss: 0.4595 - val_auc: 0.3738 - val_auc_1: 0.7190\n",
      "Epoch 5/20\n",
      "13629/13629 [==============================] - 169s 12ms/step - loss: 0.4560 - auc: 0.3669 - auc_1: 0.7190 - val_loss: 0.4593 - val_auc: 0.3738 - val_auc_1: 0.7193\n",
      "Epoch 6/20\n",
      "13629/13629 [==============================] - 174s 13ms/step - loss: 0.4556 - auc: 0.3671 - auc_1: 0.7197 - val_loss: 0.4599 - val_auc: 0.3744 - val_auc_1: 0.7207\n",
      "Epoch 7/20\n",
      "13629/13629 [==============================] - 170s 12ms/step - loss: 0.4552 - auc: 0.3689 - auc_1: 0.7206 - val_loss: 0.4583 - val_auc: 0.3757 - val_auc_1: 0.7211\n",
      "Epoch 8/20\n",
      "13629/13629 [==============================] - 2293s 168ms/step - loss: 0.4550 - auc: 0.3694 - auc_1: 0.7210 - val_loss: 0.4614 - val_auc: 0.3765 - val_auc_1: 0.7207\n",
      "Epoch 9/20\n",
      "13629/13629 [==============================] - 1114s 82ms/step - loss: 0.4548 - auc: 0.3702 - auc_1: 0.7213 - val_loss: 0.4582 - val_auc: 0.3762 - val_auc_1: 0.7214\n",
      "Epoch 10/20\n",
      "13629/13629 [==============================] - 161s 12ms/step - loss: 0.4548 - auc: 0.3694 - auc_1: 0.7213 - val_loss: 0.4587 - val_auc: 0.3781 - val_auc_1: 0.7221\n",
      "Epoch 11/20\n",
      "13629/13629 [==============================] - 162s 12ms/step - loss: 0.4545 - auc: 0.3706 - auc_1: 0.7221 - val_loss: 0.4599 - val_auc: 0.3746 - val_auc_1: 0.7209\n",
      "Epoch 12/20\n",
      "13629/13629 [==============================] - 162s 12ms/step - loss: 0.4546 - auc: 0.3715 - auc_1: 0.7217 - val_loss: 0.4602 - val_auc: 0.3784 - val_auc_1: 0.7222\n",
      "Epoch 13/20\n",
      "13629/13629 [==============================] - 2058s 151ms/step - loss: 0.4544 - auc: 0.3716 - auc_1: 0.7221 - val_loss: 0.4590 - val_auc: 0.3765 - val_auc_1: 0.7220\n",
      "Epoch 14/20\n",
      "13629/13629 [==============================] - 162s 12ms/step - loss: 0.4543 - auc: 0.3712 - auc_1: 0.7221 - val_loss: 0.4596 - val_auc: 0.3772 - val_auc_1: 0.7222\n",
      "Epoch 15/20\n",
      "13629/13629 [==============================] - 161s 12ms/step - loss: 0.4542 - auc: 0.3713 - auc_1: 0.7225 - val_loss: 0.4599 - val_auc: 0.3752 - val_auc_1: 0.7212\n",
      "Epoch 16/20\n",
      "13629/13629 [==============================] - 162s 12ms/step - loss: 0.4542 - auc: 0.3710 - auc_1: 0.7227 - val_loss: 0.4578 - val_auc: 0.3758 - val_auc_1: 0.7223\n",
      "Epoch 17/20\n",
      "13629/13629 [==============================] - 161s 12ms/step - loss: 0.4542 - auc: 0.3715 - auc_1: 0.7226 - val_loss: 0.4602 - val_auc: 0.3762 - val_auc_1: 0.7223\n",
      "Epoch 18/20\n",
      "13629/13629 [==============================] - 416s 30ms/step - loss: 0.4541 - auc: 0.3712 - auc_1: 0.7226 - val_loss: 0.4627 - val_auc: 0.3747 - val_auc_1: 0.7215\n",
      "Epoch 19/20\n",
      "13629/13629 [==============================] - 163s 12ms/step - loss: 0.4540 - auc: 0.3726 - auc_1: 0.7229 - val_loss: 0.4619 - val_auc: 0.3754 - val_auc_1: 0.7219\n",
      "Epoch 20/20\n",
      "13629/13629 [==============================] - 164s 12ms/step - loss: 0.4538 - auc: 0.3731 - auc_1: 0.7234 - val_loss: 0.4608 - val_auc: 0.3768 - val_auc_1: 0.7223\n"
     ]
    }
   ],
   "source": [
    "this_history = this_model.fit(X_train, y_train,\n",
    "                              validation_data=(X_val, y_val),\n",
    "                              batch_size=setup.nn_batch_size,\n",
    "                              epochs=setup.nn_epochs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "460c4b43-c673-4ee1-b070-d2d596ac4561",
   "metadata": {},
   "outputs": [],
   "source": [
    "this_history_df = pd.DataFrame(this_history.history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "c85bd09a-2fa9-4255-958b-4d48b17085ff",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Rename columns\n",
    "try:\n",
    "    this_history_df.columns = ['_'.join(c.split('_')[0:-1])  \\\n",
    "                               if re.search(r'_\\d+$', c) else c for c in this_history_df.columns]\n",
    "except:\n",
    "    pass\n",
    "try:\n",
    "    cur_col = list(this_history_df.columns)\n",
    "    this_history_df.columns = [cur_col[0]] + \\\n",
    "        [f'{cur_col[i]}_roc'  if (cur_col[i] == cur_col[i-1]) and 'auc'in cur_col[i] \\\n",
    "         else cur_col[i] for i in range(1, len(cur_col))]\n",
    "except:\n",
    "    pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "8ea25384-9fc0-4916-a577-4bd9bc9b203f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['loss', 'auc', 'auc_roc', 'val_loss', 'val_auc', 'val_auc_roc'], dtype='object')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "this_history_df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "ad4783f2-8d0b-4850-a45a-3a4be8e63318",
   "metadata": {},
   "outputs": [],
   "source": [
    "this_history_df.to_csv(Path(setup.temp_path).joinpath('23_REPORT_fit_history.csv'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "73e89efd-75a9-4b27-81a8-13e12d5482be",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjUAAAGdCAYAAADqsoKGAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB0bElEQVR4nO3dd3hUZfrw8e/MJJNGGoQ0CL2E3okgoEAkQemoqCxNxLKWVdYCu6tY9l10ddX9qSsuqwg2UBEsKCJgEDCKAqEIBAiE0JLQ0vvMef94MpNEEmBIpub+XNdcOTlzzpn7MCRz5yn3o9M0TUMIIYQQws3pnR2AEEIIIURDkKRGCCGEEB5BkhohhBBCeARJaoQQQgjhESSpEUIIIYRHkKRGCCGEEB5BkhohhBBCeARJaoQQQgjhEbycHYCjmM1mTp06RWBgIDqdztnhCCGEEOIKaJpGfn4+0dHR6PWXbotpNEnNqVOniImJcXYYQgghhLgKx48fp2XLlpc8ptEkNYGBgYD6RwkKCnJyNEIIIYS4Enl5ecTExFg/xy+l0SQ1li6noKAgSWqEEEIIN3MlQ0dkoLAQQgghPIIkNUIIIYTwCJLUCCGEEMIjNJoxNUIIIYSmaVRUVGAymZwdiqjG29sbg8FQ7+tIUiOEEKJRKCsr4/Tp0xQVFTk7FPE7Op2Oli1b0qRJk3pdR5IaIYQQHs9sNnP06FEMBgPR0dEYjUYpxOoiNE3jzJkznDhxgo4dO9arxUaSGiGEEB6vrKwMs9lMTEwM/v7+zg5H/E7z5s1JT0+nvLy8XkmNDBQWQgjRaFyuzL5wjoZqNZN3VwghhBAeQZIaIYQQQngESWqEEEIIF3b99dfz8MMPOzsMtyBJjRBCCCE8gsx+qqfTucUsSz6GWdOYP7qLs8MRQgghGi1pqamnwlITbyal8eFPGc4ORQghhA00TaOorMLhD03TrjrmCxcuMH36dEJDQ/H392f06NEcOnTI+vyxY8cYO3YsoaGhBAQE0K1bN77++mvruVOnTqV58+b4+fnRsWNHlixZUu9/R1ciLTX1FB3iC0B+aQX5JeUE+no7OSIhhBBXorjcRNenvnX46+57NgF/49V9/M6cOZNDhw7xxRdfEBQUxBNPPMGNN97Ivn378Pb25v7776esrIwffviBgIAA9u3bZ63S++STT7Jv3z6++eYbwsLCOHz4MMXFxQ15a04nSU09+Ru9CPbzJre4nNO5JZLUCCGEsAtLMrN161YGDx4MwAcffEBMTAyrV6/mlltuISMjg8mTJ9OjRw8A2rVrZz0/IyODPn360L9/fwDatGnj8HuwN0lqGkBUsK81qekUEejscIQQQlwBP28D+55NcMrrXo39+/fj5eVFXFycdV+zZs3o3Lkz+/fvB+Chhx7ivvvuY926dcTHxzN58mR69uwJwH333cfkyZPZsWMHo0aNYsKECdbkyFPImJoGEBWsuqBO53hWM54QQngynU6Hv9HL4Q97rjl11113ceTIEaZNm8aePXvo378/r732GgCjR4/m2LFjPPLII5w6dYqRI0fy6KOP2i0WZ5CkpgFEhfgBcCq3xMmRCCGE8FRdunShoqKCn3/+2brv3LlzpKam0rVrV+u+mJgY7r33Xj777DP+/Oc/s3jxYutzzZs3Z8aMGbz//vu8+uqr/Pe//3XoPdibdD81gKgg1VKTmSstNUIIIeyjY8eOjB8/njlz5vDWW28RGBjIvHnzaNGiBePHjwfg4YcfZvTo0XTq1IkLFy7w/fff06WLKjfy1FNP0a9fP7p160ZpaSlfffWV9TlPIS01DcDSUnNaWmqEEELY0ZIlS+jXrx9jxoxh0KBBaJrG119/jbe3mqRiMpm4//776dKlC4mJiXTq1In//Oc/ABiNRubPn0/Pnj0ZNmwYBoOB5cuXO/N2GtxVJTVvvPEGbdq0wdfXl7i4OLZt23ZF5y1fvhydTseECRMuem7//v2MGzeO4OBgAgICGDBgABkZNWu/JCcnM2LECAICAggKCmLYsGEuMR0tunJMzSkZUyOEEKKBJSUl8eqrrwIQGhrKsmXLyMnJoaioiLVr19KxY0frsa+99hqHDx+mpKSE7Oxsli1bRrNmzQD429/+xr59+ygqKuLcuXOsXr2atm3bOuOW7MbmpGbFihXMnTuXBQsWsGPHDnr16kVCQgLZ2dmXPC89PZ1HH32UoUOHXvRcWloaQ4YMITY2lqSkJHbv3s2TTz6Jr6+v9Zjk5GQSExMZNWoU27Zt45dffuGBBx5wiWXkIy0DhXNL6lVUSQghhBBXT6fZ+CkcFxfHgAEDeP311wEwm83ExMTw4IMPMm/evFrPMZlMDBs2jDvvvJPNmzeTk5PD6tWrrc/fdttteHt7895779X5utdccw033HADzz33nC3hWuXl5REcHExubi5BQUFXdY26FJeZ6PLUWgB2LRhFsJ/UqhFCCFdSUlLC0aNHadu2bY0/mIVruNT7Y8vnt03NHGVlZWzfvp34+PiqC+j1xMfHk5ycXOd5zz77LOHh4cyePfui58xmM2vWrKFTp04kJCQQHh5OXFxcjaQnOzubn3/+mfDwcAYPHkxERATXXXcdW7ZsqfM1S0tLycvLq/GwFz+jgVB/lciclsHCQgghhFPYlNScPXsWk8lEREREjf0RERFkZmbWes6WLVt4++23a0wpqy47O5uCggKef/55EhMTWbduHRMnTmTSpEls2rQJgCNHjgDw9NNPM2fOHNauXUvfvn0ZOXJkjTUvqlu4cCHBwcHWR0xMjC23arPI4MrBwjkyWFgIIYRwBrsOSMnPz2fatGksXryYsLCwWo8xm80AjB8/nkceeYTevXszb948xowZw6JFi2occ8899zBr1iz69OnDK6+8QufOnXnnnXdqve78+fPJzc21Po4fP26HO6wSXW1cjRBCCCEcz6Y6NWFhYRgMBrKysmrsz8rKIjIy8qLj09LSSE9PZ+zYsdZ9lgTFy8uL1NRUYmJi8PLyqlE4CFSRIUv3UlRUFECtx/x+hpSFj48PPj4+ttxevUSFWJIa6X4SQgghnMGmlhqj0Ui/fv3YsGGDdZ/ZbGbDhg0MGjToouNjY2PZs2cPKSkp1se4ceMYPnw4KSkpxMTEYDQaGTBgAKmpqTXOPXjwIK1btwbUolvR0dGXPMbZoiq7n05J95MQQgjhFDZXFJ47dy4zZsygf//+DBw4kFdffZXCwkJmzZoFwPTp02nRogULFy7E19eX7t271zg/JCQEoMb+xx57jClTpjBs2DCGDx/O2rVr+fLLL0lKSgLU+hyPPfYYCxYsoFevXvTu3ZulS5dy4MABPv3006u89YZlWf8pM09aaoQQQghnsDmpmTJlCmfOnOGpp54iMzOT3r17s3btWuvg4YyMDJtrx0ycOJFFixaxcOFCHnroITp37szKlSsZMmSI9ZiHH36YkpISHnnkEc6fP0+vXr347rvvaN++va23YBdRMlBYCCGEcCqb69S4K3vWqQFIP1vI9S8l4eutZ/+ziXZdhVUIIYRtGnOdmjZt2vDwww/z8MMPX/ZYnU7HqlWraq38b09OqVMj6mapKlxSbia3uNzJ0QghhBCNjyQ1DcTX20CzACMgg4WFEEIIZ5CkpgFVrQElg4WFEMLlaRqUFTr+YcOoj//+979ER0dby6FYjB8/njvvvJO0tDTGjx9PREQETZo0YcCAAaxfv77B/on27NnDiBEj8PPzo1mzZtx9990UFBRYn09KSmLgwIEEBAQQEhLCtddey7FjxwDYtWsXw4cPJzAwkKCgIPr168evv/7aYLHVxuaBwqJuUcF+/HYqTwrwCSGEOygvgn9EO/51/3IKjAFXdOgtt9zCgw8+yPfff8/IkSMBOH/+PGvXruXrr7+moKCAG2+8kf/3//4fPj4+LFu2jLFjx5KamkqrVq3qFWZhYSEJCQkMGjSIX375hezsbO666y4eeOAB3n33XSoqKpgwYQJz5szho48+oqysjG3btlnHlE6dOpU+ffrw5ptvYjAYSElJwdvbvmsjSlLTgKKlAJ8QQogGFBoayujRo/nwww+tSc2nn35KWFgYw4cPR6/X06tXL+vxzz33HKtWreKLL77ggQceqNdrf/jhh5SUlLBs2TICAlQS9vrrrzN27FheeOEFvL29yc3NZcyYMdaZyF26dLGen5GRwWOPPUZsbCwAHTt2rFc8V0KSmgZk7X6SMTVCCOH6vP1Vq4kzXtcGU6dOZc6cOfznP//Bx8eHDz74gNtuuw29Xk9BQQFPP/00a9as4fTp01RUVFBcXFxntX1b7N+/n169elkTGoBrr70Ws9lMamoqw4YNY+bMmSQkJHDDDTcQHx/Prbfeal0FYO7cudx111289957xMfHc8stt9i9DIuMqWlA0ZZaNdL9JIQQrk+nU91Ajn7YWPJj7NixaJrGmjVrOH78OJs3b2bq1KkAPProo6xatYp//OMfbN68mZSUFHr06EFZWZk9/sUusmTJEpKTkxk8eDArVqygU6dO/PTTT4BahPq3337jpptuYuPGjXTt2pVVq1bZNR5JahpQlAwUFkII0cB8fX2ZNGkSH3zwAR999BGdO3emb9++AGzdupWZM2cyceJEevToQWRkJOnp6Q3yul26dGHXrl0UFhZa923duhW9Xk/nzp2t+/r06cP8+fP58ccf6d69Ox9++KH1uU6dOvHII4+wbt06Jk2axJIlSxoktrpIUtOAoqq11DSSmoZCCCEcYOrUqaxZs4Z33nnH2koDapzKZ599RkpKCrt27eKOO+64aKZUfV7T19eXGTNmsHfvXr7//nsefPBBpk2bRkREBEePHmX+/PkkJydz7Ngx1q1bx6FDh+jSpQvFxcU88MADJCUlcezYMbZu3covv/xSY8yNPciYmgYUEaxWBS+tMHOhqJymlXVrhBBCiPoYMWIETZs2JTU1lTvuuMO6/+WXX+bOO+9k8ODBhIWF8cQTT5CXl9cgr+nv78+3337Ln/70JwYMGIC/vz+TJ0/m5Zdftj5/4MABli5dyrlz54iKiuL+++/nnnvuoaKignPnzjF9+nSysrIICwtj0qRJPPPMMw0SW11kmYQG1v/v6zlbUMpXDw6he4tgu72OEEKIK9eYl0lwB7JMgouqGlcjg4WFEEIIR5KkpoFZkppMGSwshBDChXzwwQc0adKk1ke3bt2cHV6DkDE1DSw6RA0WPiUtNUIIIVzIuHHjiIuLq/U5e1f6dRRJahpYVQE+aakRQgjhOgIDAwkMDHR2GHYl3U8NTMbUCCGE62okc2PcTkO9L5LUNDBL95MkNUII4Tos3StFRUVOjkTUxlIB2WAw1Os60v3UwCKDLAOFSzCbNfR628phCyGEaHgGg4GQkBCys7MBVWNFZ+NyBcI+zGYzZ86cwd/fHy+v+qUlktQ0sMhgX3Q6KDOZOV9URlgTH2eHJIQQAoiMjASwJjbCdej1elq1alXvRFOSmgbmbdDTvIkP2fmlnM4pkaRGCCFchE6nIyoqivDwcMrLy50djqjGaDSi19d/RIwkNXYQFexLdn4pp3KL6dFSqgoLIYQrMRgM9R67IVyTDBS2A8vClpkyWFgIIYRwGElq7CAqRA0WPiVVhYUQQgiHkaTGDqy1anKkpUYIIYRwFElq7EC6n4QQQgjHk6TGDqKl+0kIIYRwOElq7CCysqUmK08V4BNCCCGE/UlSYwcRgT7odVBu0jhbWOrscIQQQohGQZIaO/Ay6AkPlMHCQgghhCNJUmMnkdbVumVcjRBCCOEIktTYiWWwsKzWLYQQQjiGJDV2YpnWLUmNEEII4RiS1NiJpQDfqRzpfhJCCCEcQZIaO5GWGiGEEMKxJKmxE8v6T1JVWAghhHAMSWrsxNL9lJlXgkkK8AkhhBB2J0mNnYQH+mLQ6zCZNc7kSwE+IYQQwt4kqbETg15HRKAPILVqhBBCCEeQpMaOqgrwybgaIYQQwt4kqbGjqBA1A0qmdQshhBD2J0mNHUUHywwoIYQQwlEkqbEjqVUjhBBCOM5VJTVvvPEGbdq0wdfXl7i4OLZt23ZF5y1fvhydTseECRMuem7//v2MGzeO4OBgAgICGDBgABkZGRcdp2kao0ePRqfTsXr16qsJ32GsVYVloLAQQghhdzYnNStWrGDu3LksWLCAHTt20KtXLxISEsjOzr7keenp6Tz66KMMHTr0oufS0tIYMmQIsbGxJCUlsXv3bp588kl8fX0vOvbVV19Fp9PZGrZTWMbUSPeTEEIIYX82JzUvv/wyc+bMYdasWXTt2pVFixbh7+/PO++8U+c5JpOJqVOn8swzz9CuXbuLnv/rX//KjTfeyD//+U/69OlD+/btGTduHOHh4TWOS0lJ4V//+tclX8uVWMbUZOWVUGEyOzkaIYQQwrPZlNSUlZWxfft24uPjqy6g1xMfH09ycnKd5z377LOEh4cze/bsi54zm82sWbOGTp06kZCQQHh4OHFxcRd1LRUVFXHHHXfwxhtvEBkZedlYS0tLycvLq/FwtGZNfPDS6zBrkC0F+IQQQgi7simpOXv2LCaTiYiIiBr7IyIiyMzMrPWcLVu28Pbbb7N48eJan8/OzqagoIDnn3+exMRE1q1bx8SJE5k0aRKbNm2yHvfII48wePBgxo8ff0WxLly4kODgYOsjJibmCu+y4Rj0OiKCpFaNEEII4Qhe9rx4fn4+06ZNY/HixYSFhdV6jNmsumXGjx/PI488AkDv3r358ccfWbRoEddddx1ffPEFGzduZOfOnVf82vPnz2fu3LnW7/Py8pyS2ESH+HIyp7iyqnCow19fCCGEaCxsSmrCwsIwGAxkZWXV2J+VlVVrl1BaWhrp6emMHTvWus+SxHh5eZGamkpMTAxeXl507dq1xrldunRhy5YtAGzcuJG0tDRCQkJqHDN58mSGDh1KUlLSRa/t4+ODj4+PLbdnF5HBfsAFTudIS40QQghhTzYlNUajkX79+rFhwwbrtGyz2cyGDRt44IEHLjo+NjaWPXv21Nj3t7/9jfz8fP79738TExOD0WhkwIABpKam1jju4MGDtG7dGoB58+Zx11131Xi+R48evPLKKzUSJlcULUslCCGEEA5hc/fT3LlzmTFjBv3792fgwIG8+uqrFBYWMmvWLACmT59OixYtWLhwIb6+vnTv3r3G+ZbWlur7H3vsMaZMmcKwYcMYPnw4a9eu5csvv7S2wERGRtbaEtSqVSvatm1r6y04VJQ1qZFaNUIIIYQ92ZzUTJkyhTNnzvDUU0+RmZlJ7969Wbt2rXXwcEZGBnq9bTPFJ06cyKJFi1i4cCEPPfQQnTt3ZuXKlQwZMsTW8FxOZGVV4VPSUiOEEELYlU7TNM3ZQThCXl4ewcHB5ObmEhQU5LDX3X0ih3GvbyUiyIef/xJ/+ROEEEIIYWXL57es/WRnlvWfsvNLKZcCfEIIIYTdSFJjZ80CjHgbdGiaqiwshBBCCPuQpMbO9HodkZWDhWUNKCGEEMJ+JKlxgCgZLCyEEELYnSQ1DmCd1p0j07qFEEIIe5GkxgEsLTVSgE8IIYSwH0lqHCA6RArwCSGEEPYmSY0DRMpK3UIIIYTdSVLjANEh0v0khBBC2JskNQ5gGSh8tqCUsgopwCeEEELYgyQ1DtA0wIjRSy8F+IQQQgg7kqTGAXQ6XbXVuiWpEUIIIexBkhoHqUpqZAaUEEIIYQ+S1DiItapwjrTUCCGEEPYgSY2DRFnXf5KWGiGEEMIeJKlxkKgQWf9JCCGEsCdJahwkKkjG1AghhBD2JEmNg0SFWLqfpKVGCCGEsAdJahwkunKg8NmCMkorTE6ORgghhPA8ktQ4SIi/Nz5e6p9bWmuEEEKIhidJTUMpzb/k0zqdTtaAEkIIIexIkpr6Orkd3rwW3r/5sodKAT4hhBDCfrycHYDbC4yGrN8ADS4cg9DWdR4aWZnUSAE+IYQQouFJS019BUVB26Fqe+/KSx5qGSwsY2qEEEKIhidJTUPocYv6uufTSx5mmdYt3U9CCCFEw5OkpiF0GQsGI2T/VtkVVbso6X4SQggh7EaSmobgFwodR6ntS7TWWBa1lJYaIYQQF9E0KM5xdhRuTZKahtKjcvbTnk/Vf8xaWMbUXCgqp6RcCvAJIYSoZu18+Gdb+PF1Z0fitiSpaSidEsHYBHIz4Pi2Wg8J8vPCz9sASK0aIYQQ1ZQVws73QDPDur/Chufq/ANZ1E2Smobi7afG1gDs+aTWQ3Q6XdVg4RzpghJCCFEp9RsoKwDvAPX95pdgzVwwS6u+LSSpaUiWLqjfVoGpvNZDooOlqrAQQojf2b1CfR10P4x5BdDBr+/AZ3OgosypobkTSWoaUtvrwT8Mis7CkU21HhIpVYWFEEJUV3AGDm9Q2z1vhf53ws1vg95L1T9bfgeUFTk3RjchSU1DMnhB90lqu44uqGjLtG5pqRFCCAHw22egmSC6L4R1VPu6T4bbV4CXHxz+Dt6bKDOjroAkNQ3NUojvwFe1ZtZRIVJVWAghRDWWrqeeU2ru7xgP01eDTzAc/wneHQMF2Q4Pz51IUtPQWg6AkFZqwNfBtRc9XbX+k3Q/CSFEo3f2sFoYWWeoaumvrtU1MGsNBIRD1h54J0GtMyhqJUlNQ9PpoHu1mjW/IwOFhRBCWO35WH1tPwKahNd+TGQPuHOt+oP5/BF4JxGyDzguRjciSY09WLqgDq2D4gs1nrJM6c4tLqeorMLRkQkhhHAVmlZ319PvNWsPd34LzWMh/xQsSVQtPKIGSWrsIaIrhHcDczns+6LGU4E+XgQYpQCfEEI0ese3wYV0VZsm9sbLHx8UDbO+gRb91B/MS8fVOdO2sZKkxl6syybUnAWlCvBVdkHJwpZCCNF4WVppuowFY8CVnePfFKZ/Dm2vU2M3P7gZ9n9lvxjdjCQ19tJ9svqavgXyTtV4Kkpq1QghRONWUaamcoOqTWMLn0CY+olKhkxl8PE02PlBw8fohiSpsZfQ1hBzDaDB3s9qPFWV1EhLjRBCNEqH16supCYRqtXFVl4+cPO70PsPar2oz/8IyW80eJjuRpIae6qjCyrKOgNKWmqEEKJRsnQ9db9ZFW69GgYvGP86DHpAff/tX2Dj3xv1QphXldS88cYbtGnTBl9fX+Li4ti2rfZVqX9v+fLl6HQ6JkyYcNFz+/fvZ9y4cQQHBxMQEMCAAQPIyMgA4Pz58zz44IN07twZPz8/WrVqxUMPPURubu7VhO843Saq2gOnU1QtgkrRIdJSI4QQjVZJrlrAEmzvevo9nQ5G/R1GPKm+/+FF+PpRMJvrd103ZXNSs2LFCubOncuCBQvYsWMHvXr1IiEhgezsS1c5TE9P59FHH2Xo0KEXPZeWlsaQIUOIjY0lKSmJ3bt38+STT+LrW1mo7tQpTp06xUsvvcTevXt59913Wbt2LbNnz7Y1fMcKCFO1BwD2VtWsiQyWgcJCCNFo7fsCTKUQ1hmietX/ejodDHsUbvoXoINf/ger7q5zYWVPptM029qp4uLiGDBgAK+//joAZrOZmJgYHnzwQebNm1frOSaTiWHDhnHnnXeyefNmcnJyWL16tfX52267DW9vb957770rjuOTTz7hD3/4A4WFhXh5Xb7pLi8vj+DgYHJzcwkKCrri16m3XSvUf65mHeCBX0Gn41BWPje88gOBvl7seTrBcbEIIYRwvnfHQPpm1boy7NGGvfaeT2HVPWCugI4JcMu7YPRv2NdwMFs+v21qqSkrK2P79u3Ex8dXXUCvJz4+nuTk5DrPe/bZZwkPD6+1ZcVsNrNmzRo6depEQkIC4eHhxMXF1Uh6amO5uboSmtLSUvLy8mo8nCL2RrUg2bnDqhuKqvWf8ksqKCiVAnxCCNFo5J5Qs2KhqlBrQ+pxM9z2kfrcOfQtvD9ZdXc1EjYlNWfPnsVkMhEREVFjf0REBJmZmbWes2XLFt5++20WL15c6/PZ2dkUFBTw/PPPk5iYyLp165g4cSKTJk1i06baiwqdPXuW5557jrvvvrvOWBcuXEhwcLD1ERMTc4V32cB8AqHzaLVduWxCEx8vAn1VMpYpg4WFEKLx2PMpoEGrwWqWrD10GgXTVqmFMDN+hHdvajQLYdp19lN+fj7Tpk1j8eLFhIWF1XqMuXIw0/jx43nkkUfo3bs38+bNY8yYMSxatOii4/Py8rjpppvo2rUrTz/9dJ2vPX/+fHJzc62P48ePN8g9XRVLNr53JZhNQNW07lMyrkYIIRqP3ZVrPdV3gPDltB5UuRBmc8jco9aLysmw72u6AJvmkYWFhWEwGMjKyqqxPysri8jIyIuOT0tLIz09nbFjx1r3WZIYLy8vUlNTiYmJwcvLi65du9Y4t0uXLmzZsqXGvvz8fBITEwkMDGTVqlV4e3vXGauPjw8+Pj623J79dIgH3xDIPw3HtkLbYUQF+3Ewq4BMmQElhBCNQ+ZeyP4NDEboNsH+rxfZQ60XtWwCnE9Tic20VdC8s/1f20lsaqkxGo3069ePDRs2WPeZzWY2bNjAoEGDLjo+NjaWPXv2kJKSYn2MGzeO4cOHk5KSQkxMDEajkQEDBpCamlrj3IMHD9K6dVXTXF5eHqNGjcJoNPLFF19YZ0a5BS8jdB2vtitr1limdZ+S7ichhGgcLLVpOo4Cv1DHvGaz9jD7WzXTKu8kLBldo8SIp7G54s/cuXOZMWMG/fv3Z+DAgbz66qsUFhYya9YsAKZPn06LFi1YuHAhvr6+dO/evcb5ISEhADX2P/bYY0yZMoVhw4YxfPhw1q5dy5dffklSUhJQldAUFRXx/vvv1xj427x5cwwGw9Xcu2P1uAV2LIV9n8ONLxEZJNO6hRCi0TCbrOMqL7sid0OzLIT5/kQ4vQvenwR3rYcm4Y6NwwFsTmqmTJnCmTNneOqpp8jMzKR3796sXbvWOng4IyMDvd62oToTJ05k0aJFLFy4kIceeojOnTuzcuVKhgwZAsCOHTv4+eefAejQoUONc48ePUqbNm1svQ3Haz0YAqPVkvGH1xMV0hOA03mS1AghhMdL36J+//sGQycnlPIIaAZTV8Lb8Wpl8A9ugZlrwKeJ42OxI5vr1Lgrp9Wpqe7bv0Ly69BtIlt6v8Qf3v6ZjuFN+G7uVaz7IYQQwn2svh9S3od+M2Hsv50Xx7k0+F88FJ9X3WC3fXT1yzQ4iN3q1Ih6ssyCSv2GKD9Vn0aWShBCCA9XXqyGHoDju55+r1l7uOPjyjo262DNXI9aK0qSGkeK6gXNOkJFCS2z1GDrgtIK8ksaXylrIYRoNFK/gbJ8CG4FMdc4OxqIGQA3vwM6vRrr+cOLzo6owUhS40g6nbW1xmf/ZwT7qSnp0lojhBAezFqb5hawccyp3cTeCDdWJjPf/z/Y+YFz42kgLvKv24j0uFl9Tfue2ECVzJzKkWndQgjhkQrPweHv1HYPOxfcs9WAu2DII2r7y4fg8HrnxtMAJKlxtGbtIboPaCbGeKkZXVKATwghPNRvn6nFJaN6QXiss6O52IinVLJlroCPZ8CpFGdHVC+S1DhDZRfUsBK1ttUpSWqEEMIzWbuenDxAuC56PYx/A9peB2UF8OGtcOGYs6O6apLUOEO3SYCO1kV7aKnL5rR0PwkhhOc5fwRObFMDcrtPdnY0dfMywpT3ILwbFGTBBzdD0XlnR3VVJKlxhqAoaDsUgHH6ZDKlAJ8QQnie3WpZHNpdD4EXr4/oUnyDYeonENQCzh6E5XdAuft9NklS4yyVXVDjDD/KQGFhH4fWq6mkQgjH07SqtZ5ctevp94JbwNRPwScYMpJh1d1QuQi1u5Ckxlm6jEXTG4nVH6dJ7kEaSWFn4SiFZ+GjKeqvrbxTzo5GiMbn5A61Mra3P8SOcXY0Vy6iK9z2gVpJfN/nsO6vzo7IJpLUOItfKOYONwCQYN5MXkmFkwMSHuXwejWbQTPD4Q3OjkaIxsfSShN7k/utr9R2KEx4U23/9B9IfsO58dhAkhonMvSq6oI6nVPo5GiERzm0rmrbUiNDCOEYpnLYu1Jtu0vX0+/1uBlueFZtf/sX2PuZc+O5QpLUOFOnRIrwo6XuLEWHk50djfAUpoqarTNpSWqfEMIx0jZC0VnwD4N2w50dzdUb/BAMvEdtr7oH0rc6N54rIEmNM3n7sTNgCABNDq1ycjDCY5z4BUpywDcE/EKhNBdO/ursqIRoPCxdTz1udvkVsC9Jp4PEhWpMkKkMlt8O2QecHdUlSVLjZIciEgFoeWqdarIUor4sXU8d4qH9iMp90gUlhEOU5MGBNWq7p4sti3A19AaY/D9oORBKclUNm7zTzo6qTpLUOFlRyyGc1YLwr7gARzY5OxzhCSwJTMdRKrEBj1jTRQi3cOArqCiBZh0guq+zo2kY3n5wxwp1T7nH4YNbVPLmgiSpcbLIkCZ8Zapcin7PJ84NRri/vFOQtQfQQYeR0H6k2n86BQrOODMyIRqH6rVpdDrnxtKQ/JvCH1ZCQHP1O+bjaVBR5uyoLiJJjZNFBfvxhWmw+ubAV1BW5NyAhHuztNK07A8BYRAYAZE91b60jc6LS4jGIO90VYt7ZYFVjxLaBu74GLwD4EiSWtnbxWqsSVLjZFHBvuzQOnJCa64WEzu41tkhCXdmGU/TcVTVPmsXlIyrEcKu9n4KaBATB03bOjsa+2jRF25dCjoD7PoINv7d2RHVIEmNk0UG+wI6Vltaa/Z86tR4hBurKFV/PQF0vKFqvzWp2QBmk8PDEqLRsHY9ecAA4UvpeAOMfVVtb34Jfn3HqeFUJ0mNk/l6G2gWYORz07Vqx6F1UHzBuUEJ95SRrFr7AsIhslfV/piB4BMExefV2BohRMPL2geZe0DvBd0mOTsa++s7Ha6bp7bX/BlSXaOXQZIaFxAZ7MshrSX5wZ3BXA77vnB2SMIdWWc93QD6aj/aBm9od13lMTILSgi72POx+tpxlBpU2xhcPw96/0Etx/LpLDix3dkRSVLjCqKC/QA4FDFa7ZBZUOJq1DaexkKmdgthP2Yz7K78ve3pXU/V6XSqG6pDPJQXwYe3wrk0p4YkSY0LiA7xBeCXJterHelbZGVlYZvzR+HsQTV4r30tZdktSc3JX6HovGNjE8LTZfwIeSdUN2+nRGdH41gGb7hlKUT1UktDfHCzU3/HSFLjAtRgYUgtDoVWgwDNcYuHnUtTy8vLAFL3Zul6ajUIfIMvfj64JTTvopqJj3zv2NiE8HSWAcJdx6lCdY2NTxO44xMIaQUdE2r/HeQgktS4gOjK7qdTucVqrRCwfxfUye2wYhq81g8+ng7rn7bv6wn7snY93VD3MR0qC/FVX+xSCFE/5SXw2+dq211X5G4IgRFw9yYY/bxaWsFJJKlxAVGVLTWZuSXQdYLqQjidAmcPNewLaZr6QFs6FhaPgP1fAJWFk358DTJ+btjXE45RVgTpm9V2beNpLCwJz+H1LlcwSwi3dehbtWhsUAtoPcTZ0TiXCwyQlqTGBVgGCp/OLUHzb1a1CGFD1awxm2DvSnhrGLw/CY7+oKYd9rwN7kuGXrcDGqy+F8oKG+Y1heOkb1FrzQS1hPAudR/XahB4+0NBlpp6KoSov92Vs5563Fxz1qFwCnkHXEBEsA8ApRVmzheWVZXX3vNJ/f6iLi+BX95WXUyf3gmZu9WHWtx98FAKTHoLIrpC4vMQGA3nj0g3lDuydD11GnXptWa8fKDtMLUts6CEqL+i83DwW7XdmLueXIgkNS7Ax8tAWBOV2JzOLYHYG8HLD86nwamdtl+wOAc2/wte7QFr5sKFo+DXFK6fD4/8pvo8Q2KqjvcLgfGvq+1t/62qSitcn6ap5m+4dNeTRfXqwkKI+tm3WtUWi+gOEd2cHY1AkhqXYRlXczq3BHwCobOlZo0NXVB5p2Hdk/BKd9jwLBRmQ3AMJL4Aj+xVhZLq6vPsMBL636m2P38ASnLrcTfCYc4ehJwMMBirWmEuxZLUHP8JSvLsG5sQns7S9dSYatO4OElqXERVUlOsdli6oPauvPx067OHVCLy757w4/9BWT6Ed4WJb8FDO+Gae8EYcPkgbngOQlpD7nH49i/1uBvhMJaupzZDruw9btoWmrYHcwUc3WTf2ITwZBfS1dIk6KD7zc6ORlSSpMZFRIdUDRYG1F/UviFQkKkGgtbmxHZY8Qd4fQDsfA9MZWow6B0fw30/Qq/bVGGkK+XTBCa8Cehg5/sus5aHuIRLVRGuS/VZUEKIq2Mpu9F2KAS3cG4swkqSGhdhKcB3OqeypcbLCF3Hq+3qNWs0TX0YvTsG/jcC9n8JaNBpNNz5Ldy5FjolXHrA6KW0uRYG3a+2v3xIqs+6spI8OJastm1JaixdUIdkarcQV0XTqnU9yQBhVyJJjYuwdD+dsrTUQFUX1L4v1FTrPZ/CW0Ph/cmqLoneS03H/uNPcMdyaHVNwwQz4m8Q1klN/f360Ya5pmh4RzepQYpN20Oz9ld+XutrweCjyrqfSbVffEJ4qtMpajybly90GefsaEQ1ktS4CEv3U2b1pKb1YDXVujRXDf5dOVvVF/H2h2v+qKZlT1x06dokV8PbDyYsUkUA96503JINwjZX0/UEYPRXY3BAuqCEuBqWVprON4JvkHNjETVIUuMiIoOqqgqbzZVdAnoDdJ+ktovPV07L/oualp24sOa07IbWsh8Mnau21/wZ8rPs91rCdppWtd7TpZZGqIt1avd3DReTEI2BqaJqVqp0PbkcSWpcRGSwLzodlJnMnCssq3piyFzo8wcY/c/KadlPOK4U9bDHIaKHSqi+eljGX7iSzD2Qf1q12rW+1vbzLUnNsR+lirQQtjiapMpl+DWtWk9NuAxJalyEt0FP88oCfDW6oAKawfg3IO6eK5uy25C8jKp7S+8NqV/Dro8c+/qibpaup7bXgbev7eeHdVQr6prK6p5dJ4S4mKXrqftk22aXCoeQpMaFVA0WLnZyJNVEdofh89X2N09A7gnnxiMUS9dTJxvH01jodNVmQUkXVIM5lwbfPQX/ioV3EqHchX6WRf2VFlTOOEW6nlyUJDUuxLqwZY6L/SIc/Cdo0R9K8+Dz+6UbytmKzsOJbWq7w1WMp7GwjquRwcL1Ul6ixli8OwZe6wtb/626BjOSVYIjPMeBNVBeBKFtoWV/Z0cjaiFJjQuJCqmsVZNXcpkjHczgpbqhvHzVulC/vu3siBq3tI2gmVXV6PoMFm87THUtXjiqWhiEbbIPwNr58HKsmpmYvhnQqdloI55Ux2z7rxSx9CSWLvieU66+Fpiwq6tKat544w3atGmDr68vcXFxbNu27YrOW758OTqdjgkTJlz03P79+xk3bhzBwcEEBAQwYMAAMjIyrM+XlJRw//3306xZM5o0acLkyZPJyvKsGTnWpRJyXCypATUGI/5ptb3uSbWit3AO61TuerTSgFpjzFLbSFprrkxZEez8AN4eBf+Jg5/+A8UXIKglXDcPHt4DUz+BYY/CNZVFLD//I+RnOjduUX+5J6sW++11m1NDEXWzOalZsWIFc+fOZcGCBezYsYNevXqRkJBAdnb2Jc9LT0/n0UcfZejQoRc9l5aWxpAhQ4iNjSUpKYndu3fz5JNP4utbNQDykUce4csvv+STTz5h06ZNnDp1ikmTJtkavkuzdj+50pia6gbeA62HqObX1X+8/JpUouGZTVUJiK31aWojXVBX5vQu+Gou/KuzSlKO/6zqOMWOgTs+gYd3q7Fn1VvO4heo2YNF52DVvWA2Oy9+UX97PgY0tRRN07bOjkbUQadptg2QiIuLY8CAAbz++usAmM1mYmJiePDBB5k3b16t55hMJoYNG8add97J5s2bycnJYfXq1dbnb7vtNry9vXnvvfdqPT83N5fmzZvz4YcfcvPNauGwAwcO0KVLF5KTk7nmmstX0s3LyyM4OJjc3FyCglyzWNL2Y+eZ/GYyLUP92PLECGeHU7sL6fDmtVBWAKP+DoMfdHZEjcuJX+F/I8EnGB5Pq//si6zf4M3B4OUHT6Rf3UwqT1War8bKbH9XVZC1CG0DfadD76kQGHnpa5xJhbeug4piGPX/YPADdgxY2I2mwRtxcDYVxv4f9Jvh7IgaFVs+v21qqSkrK2P79u3Ex8dXXUCvJz4+nuTk5DrPe/bZZwkPD2f27NkXPWc2m1mzZg2dOnUiISGB8PBw4uLiaiQ927dvp7y8vMbrxsbG0qpVqzpft7S0lLy8vBoPVxdZ2VKTlVetAJ+rCW0DCf9PbW94To0rEI5j6XpqP7xhppOGd4XAKPWhe2xr/a/n7jRNJY6fPwAvdVb1mU6nqLFH3SbB9M/hwZ0w9M+XT2gAmneGxH+o7fVPqxYf4X5O7VQJjZcvdJvg7GjEJdiU1Jw9exaTyURERESN/REREWRm1t5nvGXLFt5++20WL15c6/PZ2dkUFBTw/PPPk5iYyLp165g4cSKTJk1i06ZNAGRmZmI0GgkJCbni1124cCHBwcHWR0yMHavvNpCIQB/0Oig3aZwtKHV2OHXrO0N1W5hKYfW9YCp3dkSNx9UujVAXna6qgNjhDQ1zTXdUfAF+fku1Qv5vpFr1vrwQmnVULZJ/PgC3LIF214Pexl77frNUN5W5HD6dLcUO3ZFlgHDsTeAb7NxYxCXZdfZTfn4+06ZNY/HixYSFhdV6jLmyn3n8+PE88sgj9O7dm3nz5jFmzBgWLVp01a89f/58cnNzrY/jx49f9bUcxcugJzywcrBwrgsOFrbQ6WDca+qH+9RO2PKKsyNqHAqy1b83VI2FaQiWaeGNbVyNpkH6VvjsblVX5pvHIfs39dd4z9tg1jfwwC+qizWg9t9fV8Ty8xIYBecOqRlTwn1UlFUti9DrdufGIi7Ly5aDw8LCMBgMF806ysrKIjLy4qbYtLQ00tPTGTt2rHWfJYnx8vIiNTWVmJgYvLy86Nq1a41zu3TpwpYtqtJpZGQkZWVl5OTk1Gitqet1AXx8fPDx8bHl9lxCZLAvmXklnM4tpldMiLPDqVtQNNz4Enw2Bza9AJ0SIaqns6PybJYieVG9ITDikofapN31atDr2VTIyVCVhj3dgTXw3QKVZFhEdFetkD1vAb/Qhn09/6Yw8S1YNh52LFWtY13HN+xruANNg6y9oNNDRDdnR3NlDq1TS8U0iYB2w50djbgMm1pqjEYj/fr1Y8OGqmZqs9nMhg0bGDRo0EXHx8bGsmfPHlJSUqyPcePGMXz4cFJSUoiJicFoNDJgwABSU1NrnHvw4EFat24NQL9+/fD29q7xuqmpqWRkZNT6uu4surJWzSlXnNb9ez1ugS5jwVyhZndUuHCXma3KitSH3osdIMVFlodo6K4nC78QaDlAbTeG1prjv8DHM1RC4x2gBv3etRHu3QJxdzd8QmPR7joY8rDa/uKhxlOdu/As7P4EPrsHXuoEi4bAW8PUIHV3YOl66nGLqtklXJrN79DcuXOZMWMG/fv3Z+DAgbz66qsUFhYya9YsAKZPn06LFi1YuHAhvr6+dO/evcb5lpaW6vsfe+wxpkyZwrBhwxg+fDhr167lyy+/JCkpCYDg4GBmz57N3Llzadq0KUFBQTz44IMMGjToimY+uRPLtO5MVyvAVxudDm56BY4lq2b7pOfVNFZ3l7YRvnpEzfQCtTxExxvq1wVRX6ZySPtebXdKaPjrd4yH4z+pcTX972z467uKwrPwyQw1viV2jCoq6RPouNcf/lc4sglO7VAf8jO+AL3Bca/vCKZyOPGLSpAPb6gcHP27iQ/mCvh5keqWc2VF5+Hgt2q79x3OjUVcEZuTmilTpnDmzBmeeuopMjMz6d27N2vXrrUOHs7IyEBv40C6iRMnsmjRIhYuXMhDDz1E586dWblyJUOGDLEe88orr6DX65k8eTKlpaUkJCTwn//8x9bwXZ51/SdXWyqhLk2aw5hX4ONpsPVV6HwjxAxwdlRXp/AsfPtX2L1cfR/UArz94Nxh2Ph3GPuq82I7vg1Kc8G/GUT3afjrd4hX93gkSY0h8DI2/Gs4m9mkKv/mnVQDgCe86diEBtSMtcn/g0VD4dgW9TMz9M+OjcEeLqSrBCZtIxz9QS2pUl1kD2g/UnW7aRosG6cWhox/RnXNuaq9K1UCHNnDfbrLGjmb69S4K3eoUwOwZvdp7v9wB/1ah7LyvsHODufKfXY37F4BzTrAPZvB6O/siK6cpsGu5fDtX1TfOTq1KvqIv0HmHlgyWo0BuGezWuDTGb5boD4Ae06BSf9t+OubzfCvTlB4BmZ8BW0vLpLp9jY8B5tfAm9/mLMRwrs4L5aUD2H1fWos0+x17reOUFmhWt398AZI26AS/+r8m0H7ESqRaT+i5hgwTVPdT5m7VVJj6ZJzRf8drlrVEhbCoD86O5pGy251aoT9WdZ/ynTl2U+1Gf1C5eyOw7DhWWdHc+XOH4H3Jqip6cXnIbwb3LVe3Y9PILQeDF0nqLWW1s5z3mKe9hpPY6HXqw8g8MxxNalrVUIDqsvDmQkNqFk03SeDVtl6VJrv3HguR9Mgc69arHPpOHihDXx4K2x7S/3M6wzQarD6Q2DO9/DoYdUi1fv2iwe163QQd6/a/uV/YKpw+O1ckTOpKqHRGaDHzc6ORlwhSWpcTHS1MTUmVy3AVxu/UBinqkzz85uqCdqVmcrVVPT/DFJdLl6+MHIB3LPp4r+ab3gWDD5qwcIDXzk+1pzjkL1PtRa1t2OlaU9dMuH8UVh1t9oeeI9rfEDpdHDTyxDcSnXdfP2YsyO6WNF5NZV59R/h5S6w6Fq16vjRTWAqU7Pk+s2CKe/DE0fhzm9g2GPQou/la/l0n6xac3KPw8FvHHM/trIMEO54AzQJd24s4opJUuNimgf6YNDrMJk1zuS72WyijvHQb6baXn2/6/71eWI7/Pd6VeG1ogTaXgf3/QhD59ZepTe0ddVyEOv+5vhZXocrp3K3HGjf8QftRwA6NeU277T9XseRyovVeK+SXPXvN+rvzo6oil+I6krU6dUH6O5PnB2RcmANLB4B/2ynWpFSPoD806rbruMoSHwBHvgV/rRbjTPrMtb2gnTevlW/K35+q6HvoP7MJjXmB2TxSjcjSY2LMeh1RASq+jouu7DlpYz6u/oLLjdDDbp1JaX5aibT/0aqD26/pjBhkSp936z9pc8d8gg0iVR/Vf/k4AHqlvo09V2V+3ICmqm/skGNk3B3mgZrHlXjovzD4JZ3XW8AdOtBMOxxtb1mbtWMO2coLVDLQyy/A05uBzTVHTv4QZi2Gh4/qlYgv+ZeCOuoWpvqo/9s1bWTvtn1pncf/UENKPcNhk6jnR2NsIEkNS4oKsSyWrebjasBNQ5lwptqe8fSqg9kZ0v9Ri1I9/MiQFMVYx/4RfX5X8kvZ58mEP+02v7hJcjPuuThDaaiVHWPgf3G01Rn6YJylfetPnYsg5T3VUvIze9AcAtnR1S7YY9BTJyaMbRyjnPGmJz4VdWP2fkeoIPBD8Hc/fDHH9UfKu2HN/xip8EtVCsPuF5rza7KGZDdJskir25GkhoXFOlu07p/r80QuKZypsDnD6i+eWfJz4SPp8NHt6m/vELbwLRVMOkt2+vO9JwC0X3VCuUbn7NLuBc5thXKi1QrUWQP+7+eJak58r3rDuC8Eqd2Vo1TGfGkKnznqgxeMGkx+ATBiW3wwz8d99qmCkh6Ad4eBReOQlBLmPkVjHpOVQ23t7h71NfdHzv390R1pfmw/wu1LbVp3I4kNS4oOtgN1n+6nJFPqVogBZlq6uqRTY79pWU2w6/vwOsDYd/nqpn72ofhvuSrH2yr10Pi82p75/uOWXH5oGXW0w31b+6/Ei36gW+IGoNycrv9X88eis6rRNZUquomXfuwsyO6vNDWqt4TwA8vqoKW9nb+iCpXkPQPNQur+81w31b1R4mjtBqkkvWK4spWIhew/0v1h0TT9lWVtoXbkKTGBVmrCrtzUuPtp6q16vRwcK0qtvXPtvBKD/joDlV9+MDXqlR8Q0+TPpMK796oqgKX5qpidXcnwQ3P1L9+Tqs49csfDb5xwBRve0/l/j29oSrpO+yGXVBms6qZlJMBoW1VV6itq2o7S4+bodcdqnzAZ3OgOMc+r6NpKilfNFS1DPkEwaT/wc1vq8HLjqTTqRlpANv+pwboOlvKh+prryvsmhYuxU1+2hsXa1VhdxwoXF3L/nD7ctVvHqLW8SI3A1LXQNJCWH47vNJNzbJYNh7WPalmgJxJvbpfbhWl8P1CePNayEhW6/okPg93bWjYxTZveAa8/CDjR9i3uuGu+3vn0uB8Gui91aKTjuLOU7t/eFElY16+MOU9x39I19eN/1TJWO5x+Orhhk+aLa1Yn9+vulFbX6taZ3re0rCvY4seN6tB+7kZauybM+VkqIHLAD1vdW4s4qrI6lwuyDpQ2B0WtbycTglVaxUV56hZR6d3q2qip3fDmQOq6N2RpKoBsaCmj0Z0g8ieKiGJ7AnhXesetHfsR/jyT3D2oPq+YwLc9C8IiWn4ewpuCdf+CTY9D+ueUiuUe/s1/OtYBuu2HgS+DqyC3aGyCN+pnVBwRi2F4Q4Or1fJMqiuHEeMQWpoPoEw+W14ZxT8tgo63AB9pjbMtdM2qpoz+adB76XWobr2T85fe8rbD/rNUHWjtr0FXcY4L5bdK9TXNkNVl6BwO5LUuCDLmJrs/BIqTGa8DB7SoOYXovrrq/fZl5eownKZu9XU29O7VeJTXqQWxTvxS9WxOgM071wz0QltDZv/BdvfVccEhKu/drtOsG/T8bV/UmMAcjMg+XU1g6WhObrrySKwclBy5h71QdhrimNf/2rkZMDKuwBNFYRz5wGeLfvB8L+oytxfPwatrrl8yYFLKS+BDc9UlSJo1lFV+43u3SDhNoj+s2Hr/6mp1Fn7IKKr42OwLJcCUpvGjUlS44KaNfHBS6+jwqyRnV9KdIgdWgFchbevqo1iqY8CquvpXFpla86uqlad4vMqAcreV7XoZHV9Z6iuIb9Q+8dt9Ffr1nx2F2x+BXr/AYKiGu76lrV1wPFJDaguqMw9qvXD1ZOailL4eAYUX1DjpyyDud3ZtQ+rVdnTN6sCeHeuu7oaO1m/qWni2ZV1YAbcBTc853prs4XEQOxNatbRtv86Z/HYE7+qJR+8/KDreMe/vmgQHtIE4FkMeh0RQZYZUG4+ruZq6A3QvJPqax/1nCqO9/gReOQ3NUbn+r9A7BhVYh4grDPM/BrG/Z9jEhqLHjerKrXlheov4YZ09Ac1eyekFYR1athrX4kOlYX+0jaowbeubO08tUaPXyjcuswz6oroDTDxLTUT7dRO+P7/2Xa+2QzJb6jK2dm/QUBzuONj1SXragmNhWU9qN0rVILqaJZlEbqMdfzq7aLBSEuNi4oO8eVkTrF7T+tuSDqdGssS3BI6V6vwWVaoxt84Y5aCTgejn1cl5Xd9BAPmqK6DhmDtekpwzr3FDARjIBSdg9M71VRvV5TykZq6j07N4Alp5eyIGk5wC7X45sfT1EKS7UdcWb2dvFOVZRSS1PedEtW6bK4+Nqr1YIjorrqfd75ftTSJI1SUwt6Varv37Y57XdHgpKXGRUUGe9BgYXsyBjh32mWLfmrqJzTcKt6aVm1pBCd0PYFaA8vyAXrYRZdMyNyjZggBXD9PrT3mabqOq1wjSYNV91y+1tO+z+HNwZWLtPqpAdO3L3f9hAYqV++2TO/+r2Ondx9cCyU5EBit1oITbkuSGhcV7SnTuhuDkQvU9PET29SqxvV15oCa0uvl69hCaL9nWWvKFad2F+fAimlqQdION1Stn+SJEv6huiDzT8MXD9aeOJfmq5lNH09XXTdRveHezdD/TveqtdLjFtWNmJMBB7913OumVHY99bzV+bPBRL1IUuOiLLVq3LoAX2MRFAVDH1Hb6xeoLrH6sHQ9tRnq3PEP7Sundp/4xXVK2IMaL7L6PlXWP7iVWunaXQrsXQ1jgJqtZDDCga9g+5Kaz2f8rNZtSvkA0MHQP8Ps79Sik+7G2w/6TlfbPy9yzGsWnKkqNNlLup7cnQf/JnBvlu6nU5LUuIdBD6gP2LyTampqfTi768kiJAaax6oKt9VrCDnbj/+G1K/Vh/yUZeDf1NkR2V9Ur6oFVdf+BbIPgKkcvv8HLElUq3sHt4JZX6slSlxtNXJbDLhLVSI/ukndp73t/RTMFWrmXHis/V9P2JUkNS4qOqRy9pO7LmrZ2Hj7qenkoAZ15p64uuuU5KpqyFDV/eNM1urCLjKu5sgmVb8F4MYX1QdRYxF3n2o9qyiGT++EdxJh0wsq6ew5Be7bogbburuQVmrNLlDF+OzNMutJWmk8giQ1Lsqy/tOZglLKTS4+pVYo3SZCq8HqQ2f901d3jbTv1V+NYZ2gadsGDe+qVF8ywd7rXF1O3in1Ya6ZofdUVZeoMdHr1VpW/mFqmvbJX8E3WFUgnvRfte0pLNO7dy233xpYoAr9nd6lKix3v9l+ryMcRpIaF9UswIi3QYemQVaedEG5BZ0OEhcCOtjzCRzfZvs1XKXryaL1YDVlviBTTbV1looyVWCv6CxE9FD1VtxpAGxDCYxQ9Wu8/dUsnft+VPWSPE2bIWpZlPIiNb3bXiytNB0TIKCZ/V5HOIwkNS5Kr9cRGWwpwCdJjduI7l21Vs83T9hWuM5srhqw6ApdTwBePtB2mNp25iyo756sXFE6WI2jscdaW+6iYzw8cQxmfKHqNnmi6tO7f1lsn+ndpgrY/bHalto0HkOSGhdm6YKSpMbNjHhKFa47taNqgbwrkbkbCrLA2ARaDbJffLaydEEdclJSs+fTqpkwk96Cpu2cE4crceeBwFeqx62qovKF9KoZgQ3paJJqgfQLdZ2WUVFvktS4MMu0bhks7GYCI2DYn9X2+qehtODKzrP84m53vWohcRWWVbuP/wQleY597ewD8MVDanvon2tWkxaezehfbXq3HQYMW2rTdL/ZtX7eRL1IUuPCpKXGjV3zRwhto/4S3PLKlZ3jrFW5L6dpO2jaXg1gPvqD4163NB9W/EGtrdX2Ohj+V8e9tnANlundR76HM6kNd92SPFXzB2TWk4eRpMaFWad1S1Vh9+PlA6P+rrZ/fA0uHLv08YXn1CrB4DrjaaqzzoL6zjGvp2nw+f1w7hAEtYCb35FKr41RaOtq07v/23DX3bdaVaMO6wQt+jbcdYXTSVLjwiKDZKCwW4sdo6oCm0rhu6cufWzaBkBTM3uCoh0Snk2q16ux99TuvFPw2Ry1jpHeG25ZCgFh9n1N4boG3q2+pnyk6jg1hF3L1ddetzXOWXQeTJIaFxYdUllVWBa1dE86HSQ+r5rP962G9K11H2vtenLBVhpQU2wNPmpNqrMH7fMaZUWQ9Dy81k9NiQe48Z8QM8A+ryfcQ9th0LyL6obc+UH9r3chHY5tBXSqaKHwKJLUuDDLQOGzBaWUVUgBPrcU2b2qSNzaebVPTTWbqqZLu9p4GgujP7S5Vm0fauAuKLMZdq1QyUzSQlWbJOYamLNRLcgoGjedDuIqW2u2/de2Mgm12VU5I7HtMM+dEt+ISVLjwpoGGDF6qbdICvC5sRF/A58gNWU7pZa/NE/8qlZW9g2Bli7cKlG9unBDyfgZ/jcSVt0N+afU+kU3L4E710KLfg33OsK99ZyiKiZfOFq/cV2aVlVwr/cdDRObcCmS1LgwnU5nba05JdO63VdAGFz3uNre8OzF06ItXU8dRoLBy7Gx2aJDZdfYsa31X4k8JwM+mQXvjFL1fIxNYOQCeOAX6D5JxjmImowB0Gea2q7P6t3Hf1aJkXeAGvMmPI4kNS7OktRkSkuNext4j5oWXXgGNr9U8zlXncr9e2EdVUuKqQzSt1zdNUrzYf0z8Fp/+O0zQKdqkTy4A4bOBW/fBg1ZeJCBcwAdpG2EM1c5rsvSStN1PPg0abDQhOuQpMbFWWrVyGBhN+dlhIT/p7Z/ehPOH1HbeadVtxQ6tQKzK9Ppqgrx2doFZTbBjmXwf31hy8tqRliboXDvZhj3mipYKMSlhLapKr54NdO7y4th7yq13eu2BgtLuBZJalyctaqw1Kpxf50Sod1w1dKx7km1z5IctOgLTZo7L7YrZZmdZUtSc/QH+O918MWDUJitivnd9iHM+BIie9gnTuGZLOtB7bqK6d2p30BpLgS1VAm18EiS1Li4NmEBAGzYn01xmR0WdROOY1nFW2dQ1UyPbHKfrieLtsNA76Vams6lXfrYc2mwfCosHQuZe9RilAn/gD/+DLE3ybgZYbu210HzWCgrgJQPbTvX0vXUawro5aPPU8k76+LG9oymRYgfJ3OKeXPTZT5EhOsL71I1TXntfEj7Xm27S1LjE1i12ObhDbUfU5wD3/4V3ohTyZvOoMrdP7QTBt3fOBZjFPah01UV47Nlend+VtX/V1kWwaNJUuPi/IwG/npTFwAWbUrj+PkiJ0ck6m34X9T07ezfoCwfAppDVG9nR3Xl6prabaqAX/4Hr/WF5NfBXK6Ove9HuOlfENDM8bEKz9Nzimr1O3/kyrtB93wCmgla9FcD3oXHkqTGDYzuHsng9s0oqzDz3Ff7nB2OqC//pnD9/KrvO9zgXs3hlqTm6A9QXjmA/fB6WDQE1vwZis5BWGeY+in8YSWExzovVuF5fJpA38rp3duucPVuy7IIvaWVxtO50W/Sxkun0/H0uG4Y9DrW7cvih4NnnB2SqK8Bs9UHP6jxJe4kohsERkFFsZrR9MEt8P5kOLMf/JrCjS/BfVtdd8kH4f4G3AXoVDJ99tClj83cA1l7wGCEbpMcEp5wHklq3ESniEBmDGoDwNNf/ibLJrg7gzfM+AKmfOB+SU31qd3fPKYGO+u94Jr74aEdqp6Iwdu5MQrP1rStmk0IsG3xpY+1tNJ0SlStpMKjSVLjRh6+oSNhTYwcOVPIuz8edXY4or4CI6HLGPecBWT5QAHofBPcvw0S/wF+oc6LSTQulvWgUj64uEq3hakCdn+stmWAcKMgSY0bCfL15vFENT7h3+sPkS1VhoWzxI6BCW/CzK/h9g+hWXtnRyQam3bDVRduWUHVdO3fS9uoaiP5N5Pu0EbiqpKaN954gzZt2uDr60tcXBzbtm27ovOWL1+OTqdjwoQJNfbPnDkTnU5X45GYmFjjmIMHDzJ+/HjCwsIICgpiyJAhfP/991cTvlu7uW9LeseEUFhm4vlvDjg7HNFY6XRqQUDLyt1COJpOV7l0AvDzW7VP795VWcumxy3SJdpI2JzUrFixgrlz57JgwQJ27NhBr169SEhIIDs7+5Lnpaen8+ijjzJ0aO2VHBMTEzl9+rT18dFHNTPvMWPGUFFRwcaNG9m+fTu9evVizJgxZGZm2noLbk2v1/HMuG7odPDZzpNsP3be2SEJIYRz9LodfILgfJpqlamuOAcOfF11nGgUbE5qXn75ZebMmcOsWbPo2rUrixYtwt/fn3feeafOc0wmE1OnTuWZZ56hXbt2tR7j4+NDZGSk9REaWtU3f/bsWQ4dOsS8efPo2bMnHTt25Pnnn6eoqIi9e/faegtur1dMCLf2iwHgqc9/w2TWnByREEI4gU8T6PMHtf371bt/W6XWGGveBaJ6OT424RQ2JTVlZWVs376d+Pj4qgvo9cTHx5OcnFznec8++yzh4eHMnj27zmOSkpIIDw+nc+fO3HfffZw7d876XLNmzejcuTPLli2jsLCQiooK3nrrLcLDw+nXr1+t1ystLSUvL6/Gw5M8ltiZQF8vfjuVx/JfMpwdjhBCOId1evd3NZfuqF6bxh0H44urYlNSc/bsWUwmExERNVfUjYiIqLMbaMuWLbz99tssXlz3tLvExESWLVvGhg0beOGFF9i0aROjR4/GZFJrHel0OtavX8/OnTsJDAzE19eXl19+mbVr19Zo0alu4cKFBAcHWx8xMTG23KrLC2viw9wbOgHw0rep5BSVOTkiIYRwgmbtq5YZsazefS4Njv8EOj30uNV5sQmHs+vsp/z8fKZNm8bixYsJCwur87jbbruNcePG0aNHDyZMmMBXX33FL7/8QlJSEgCapnH//fcTHh7O5s2b2bZtGxMmTGDs2LGcPn261mvOnz+f3Nxc6+P48eP2uEWn+sM1rekU0YQLReX8a91BZ4cjhBDOYZnevfMDKM2H3SvU9+2GQ1CU8+ISDmdTUhMWFobBYCArK6vG/qysLCIjIy86Pi0tjfT0dMaOHYuXlxdeXl4sW7aML774Ai8vL9LSal+gsV27doSFhXH48GEANm7cyFdffcXy5cu59tpr6du3L//5z3/w8/Nj6dKltV7Dx8eHoKCgGg9P423Q8/S4bgB88PMx9p3yrC42IYS4Iu1GQLOOai21lA+rrcgtA4QbG5uSGqPRSL9+/diwoWp1XrPZzIYNGxg0aNBFx8fGxrJnzx5SUlKsj3HjxjF8+HBSUlLq7BI6ceIE586dIypKZdhFRWoRR/3v1sfR6/WYr3SVVg81uH0YN/WIwqzB01/8hqbJoGEhRCOj11et3r3x75CTAcZA96vWLerN5u6nuXPnsnjxYpYuXcr+/fu57777KCwsZNasWQBMnz6d+fPVYn2+vr507969xiMkJITAwEC6d++O0WikoKCAxx57jJ9++on09HQ2bNjA+PHj6dChAwkJCQAMGjSI0NBQZsyYwa5duzh48CCPPfYYR48e5aab5D/tX27qgq+3nm3p5/li1ylnhyOEEI7X+3aVyJRWtlh3Gw9Gf+fGJBzO5qRmypQpvPTSSzz11FP07t2blJQU1q5dax08nJGRUec4l9oYDAZ2797NuHHj6NSpE7Nnz6Zfv35s3rwZHx8fQHV7rV27loKCAkaMGEH//v3ZsmULn3/+Ob16yVS9FiF+3H99BwD+8fV+CksrnByREEI4mE8g9Jla9X2vO5wXi3AandZI+ivy8vIIDg4mNzfXI8fXlJSbGPXKD2ScL+Le69ozb3Sss0MSQgjHOpcGi4aqBS/v2ay6pYTbs+XzW95xD+HrbeDJMV0BeHvLEY6cKXByREII4WDN2sOD22HW15LQNFLyrnuQ+C7hXNepOeUmjWe/2ieDhoUQjU9QFPgGOzsK4SSS1HgQnU7HgrFd8TboSEo9w4b9l16PSwghhPAkktR4mHbNm3DnkLYAPPvVPkrKTU6OSAghhHAMSWo80IMjOhIe6EPG+SL+t/mIs8MRQgghHEKSGg/UxMeLv9zYBYA3vk/jVE6xkyMSQggh7E+SGg81vnc0A9qEUlxu4h9f73d2OEIIIYTdSVLjoXQ6HU+P64ZeB1/tPk1y2jlnhySEEELYlSQ1HqxbdDB3xLUC4Jkvf6PC1LjXyRJCCOHZJKnxcH++oTMh/t4cyMzn/Z+OOTscIYQQwm4kqfFwoQFGHh3VGYCXvzvIuYJSJ0ckhBBC2IckNY3A7QNb0S06iLySCl78NtXZ4QghhBB2IUlNI2DQ63hmXDcAVvx6nN0ncpwbkBBCCGEHktQ0Ev3bNGVinxZoGjz1+W+YzbIulBBCCM8iSU0jMn90LAFGAynHc1i544SzwxFCCCEalCQ1jUh4kC8PjewIwAtrD5BXUu7kiIQQQoiGI0lNIzPr2ra0CwvgbEEZ/15/yNnhCCGEEA1GkppGxuil56mxXQFY+mM6h7LynRyREEII0TAkqWmEru8cTnyXCCrMGk9/+RuaJoOGhRBCuD9Jahqpp8Z0xeilZ+vhc6zdm+nscIQQQoh6k6SmkWrVzJ97hrUD4O9r9lNYWuHkiIQQQoj6kaSmEfvj9R2IDvblZE4xN/3fZrYdPe/skIQQQoirJklNI+ZnNPDaHX2ICPIh/VwRt76VzILP90qrjRBCCLckSU0j1691U9Y9ch1T+scAsDT5GAmv/sDWw2edHJkQQghhG0lqBMF+3rxwc0/emz2QFiF+nLhQzNT//cz8z3ZLgT4hhBBuQ5IaYTW0Y3O+fWQY0we1BuCjbcdJeOUHvk/NdnJkQgghxOVJUiNqaOLjxbPju7P87mto3cyf07klzFryC3M/TiGnqMzZ4QkhhBB1kqRG1Oqads1Y+6dh3DWkLTodfLbjJDe88gPf/iY1bYQQQrgmSWpEnfyMBv42piuf3juY9s0DOJNfyj3vbeeBD3dwrqDU2eEJIYQQNUhSIy6rX+tQ1jw0lPuub49Br+Or3ae54ZUf+HLXKVliQQghhMuQpEZcEV9vA08kxrLqj4OJjQzkfGEZD360k3ve2052XomzwxNCCCEkqRG26dkyhC8eGMLD8R3x0utYty+LG175gZXbT0irjRBCCKeSpEbYzOil5+H4Tnz54BC6twgit7icP3+yizvf/YXTucXODk8IIUQjJUmNuGpdooJY/cdreTyxM0aDnu9TzzDq5R/4aFuGtNoIIYRwOElqRL14GfT88foOfP2nIfRpFUJ+aQXzP9vDH97+mePni5wdnhBCiEZEkhrRIDqEB/LpvYP5201d8PXWs/XwORJe/YGlP6ZjNkurjRBCCPvTaY2knyAvL4/g4GByc3MJCgpydjgeLf1sIY+v3M22o+cB6NMqhHuGteeGrhEY9DonRyeEEMKd2PL5LUmNsAuzWeODn4+x8JsDFJWZAGgR4sf0Qa2ZMiCGEH+jkyMUQgjhDiSpqYUkNc6RlVfCsuR0Pvw5gwtFasVvX289E/u0ZObgNnSODHRyhEIIIVyZJDW1kKTGuUrKTXyx6xTvbk1n3+k86/5B7Zox89o2xHeRrikhhBAXk6SmFpLUuAZN0/gl/QJLf0xn7W+ZmCoHEbcMreya6t+KYH9vJ0cphBDCVUhSUwtJalzPqZxi3v/pGB9tq+qa8vM2MLFvC2YObkOnCOmaEkKIxs6Wz++rmtL9xhtv0KZNG3x9fYmLi2Pbtm1XdN7y5cvR6XRMmDChxv6ZM2ei0+lqPBITEy86f82aNcTFxeHn50doaOhF1xHuJTrEj8cTY0meP5J/Tu5Jl6ggistNfPhzBqNe+YGp//uJ7/ZlWVtzhBBCiEvxsvWEFStWMHfuXBYtWkRcXByvvvoqCQkJpKamEh4eXud56enpPProowwdOrTW5xMTE1myZIn1ex8fnxrPr1y5kjlz5vCPf/yDESNGUFFRwd69e20NX7ggX28Dtw6I4Zb+Ldl29Dzv/pjOt79lsvXwObYePkdMUz9mDGrDLf1jCPaTrikhhBC1s7n7KS4ujgEDBvD6668DYDabiYmJ4cEHH2TevHm1nmMymRg2bBh33nknmzdvJicnh9WrV1ufnzlz5kX7qquoqKBNmzY888wzzJ4925ZwraT7yb2czCnmveRjLP8lg5xqXVOTKrumOkrXlBBCNAp2634qKytj+/btxMfHV11Aryc+Pp7k5OQ6z3v22WcJDw+/ZEKSlJREeHg4nTt35r777uPcuXPW53bs2MHJkyfR6/X06dOHqKgoRo8efcmWmtLSUvLy8mo8hPtoEeLHvNGx/DR/JC9M7kFsZCDF5SY++DmDG175gT/872fWS9eUEEKIamxKas6ePYvJZCIiIqLG/oiICDIzM2s9Z8uWLbz99tssXry4zusmJiaybNkyNmzYwAsvvMCmTZsYPXo0JpMq2nbkyBEAnn76af72t7/x1VdfERoayvXXX8/58+drvebChQsJDg62PmJiYmy5VeEifL0NTBnQim/+NJTld19DYrdI9DrYcvgsdy37leEvJfG/zUfIKyl3dqhCCCGczOYxNbbIz89n2rRpLF68mLCwsDqPu+2226zbPXr0oGfPnrRv356kpCRGjhyJ2WwG4K9//SuTJ08GYMmSJbRs2ZJPPvmEe+6556Jrzp8/n7lz51q/z8vLk8TGjel0Oq5p14xr2jXjxIUi3vvpGMu3HSfjfBF/X7Ofl787yM39WjJjcBvaN2/i7HCFEEI4gU1JTVhYGAaDgaysrBr7s7KyiIyMvOj4tLQ00tPTGTt2rHWfJUHx8vIiNTWV9u3bX3Reu3btCAsL4/Dhw4wcOZKoqCgAunbtaj3Gx8eHdu3akZGRUWusPj4+Fw02Fp6hZag/80d34eGRnVidcpIlW49yMKuAZcnHWJZ8jOs6NWfWtW0Y1rE5einoJ4QQjYZN3U9Go5F+/fqxYcMG6z6z2cyGDRsYNGjQRcfHxsayZ88eUlJSrI9x48YxfPhwUlJS6mw5OXHiBOfOnbMmM/369cPHx4fU1FTrMeXl5aSnp9O6dWtbbkF4ED+jgdsHtuLbh4fx4V1xxHeJQKeDTQfPMHPJL8S/sollyekUlFY4O1QhhBAOYHP309y5c5kxYwb9+/dn4MCBvPrqqxQWFjJr1iwApk+fTosWLVi4cCG+vr507969xvkhISEA1v0FBQU888wzTJ48mcjISNLS0nj88cfp0KEDCQkJAAQFBXHvvfeyYMECYmJiaN26NS+++CIAt9xyy1XfvPAMOp2OwR3CGNwhjIxzRSxNTufjX45z5EwhT33+Gy+uTeXWATHMGNSGVs38nR2uEEIIO7E5qZkyZQpnzpzhqaeeIjMzk969e7N27Vrr4OGMjAz0+itvADIYDOzevZulS5eSk5NDdHQ0o0aN4rnnnqvRffTiiy/i5eXFtGnTKC4uJi4ujo0bNxIaGmrrLQgP1qqZP0+O6crcGzqxcscJ3t2azpGzhby95SjvbD3KyNgI7ry2DYPaN0Onk64pIYTwJLJMgvBoZrPGD4fOsGRrOpsOnrHu7xwRyMxr2zChdwv8jAYnRiiEEOJSZO2nWkhSI9LOFLD0x3Q+3X6CojJVLiDE35vbBrRi2qDWtAjxc3KEQgghfk+SmlpIUiMscovL+eTX4yxNTuf4+WIADHodCd0imDm4LQPahErXlBBCuAhJamohSY34PZNZY+OBbJZsPcqPaVUVrLtFBzFzcBvG9orG11u6poQQwpkkqamFJDXiUg5k5rH0x3Q+23GS0gpVS6lZgJGpca24dUAMLUNl1pQQQjiDJDW1kKRGXIkLhWUs/+U47yWncyq3xLp/ULtmTO7XktHdIwnwsWshbiGEENVIUlMLSWqELSpMZtbty+K95GMkH6nqmvI3GhjdPYrJ/VpwTdtmUrFYCCHsTJKaWkhSI67WiQtFrNpxkpU7TpB+rsi6v0WIH5P6tmBy35a0CQtwYoRCCOG5JKmphSQ1or40TWNHxgU+3X6Sr3adIr/a8gv9WocyuW9LbuoZRbCftxOjFEIIzyJJTS0kqRENqaTcxLp9WazcfoLNh85grvwpMnrpGdU1gsn9WjK0QxheBpuWVxNCCPE7ktTUQpIaYS9ZeSWs3qm6pw5mFVj3hwf6MLFPCyb3a0mniEAnRiiEEO5LkppaSFIj7E3TNPaezGPljhN8nnKSC0Xl1ud6tAhmct8WjOvdgqYBRidGKYQQ7kWSmlpIUiMcqazCzPep2azcfoKNB7KpqOyf8jboGN45nMn9WjK8czhGL+meEkKIS5GkphaS1AhnOVdQyhe7TrFyxwn2nsyz7m8aYOTGHpEMaNOU3jEhtGrqL8szCCHE70hSUwtJaoQrSM3MZ+WOE6zaeZIz+aU1ngvx96ZnyxB6tQymV8sQesYEEx7o66RIhRDCNUhSUwtJaoQrqTCZ2Xz4LEkHstl1Ipd9p/IoM5kvOi462JdeMSEq2YkJpkeLYAJ9Zcq4EKLxkKSmFpLUCFdWVmHmQGYeu07ksut4DruO53D4TAG//+nU6aB98yb0qkxyerUMITYqEB8vWXhTCOGZJKmphSQ1wt0UlFaw50Quu0/ksOtEDruO53Iyp/ii44wGPV2iAq0tOr1jgmkX1kSWcBBCeARJamohSY3wBGfySyuTnMoWnRM55FSbOm4R6ONF9xbBDGgTSnzXCLpHB0uSI4RwS5LU1EKSGuGJNE3j+PliUk6oLqvdJ3LYczKXkvKa43MignyI7xJBfNcIBrVrhq+3dFcJIdyDJDW1kKRGNBYVJjOHsgtIOZ7DDwfPsOngGYrKTNbnA4wGhnVqTnyXCEbEhhMqxQCFEC5MkppaSFIjGquSchPJR86xfl8W6/dnkZVXNZVcr4P+bZoyqmsE8V0iZLVxIYTLkaSmFpLUCAFms8beU7ms35fFun1ZHMjMr/F8h/Am3FCZ4PSJCZFxOEIIp5OkphaS1AhxsePni1i/X7Xg/HzkvHU5B4CwJkZGxqpxOEM6hOFnlHE4QgjHk6SmFpLUCHFpucXlJKVms35/NkkHsskvrbA+5+utZ0iH5ozqGsHw2HCaB/pc1WtomkZphZmC0goKSysqv5ooLK2gsMyyT31vMmsM6xRG31ahsnyEEI2YJDW1kKRGiCtXVmFm29HzrN+fxXf7smrUx9HpoE9MCPFdIwgL8KlKUMoqKKpMSAoqkxRLgmJ9lJkwmW37ldOmmT+T+rZkYp8WxDT1b+hbFUK4OElqaiFJjRBXR9M09p/O57vKgcZ7TuY2yHX9jQYCfLwIsHz18aKJ9auBglITG/Zn1Zi5NbBtU27u25LRPSJluQghGglJamohSY0QDeN0bjHr92fzw8EzVJjMNZKR2hKUAB8DAcbq+wz4G70wXMEg5MLSCtbuzeSznSf4Me2cddkIX289Cd0imdS3JUM6hF3RtYQQ7kmSmlpIUiOEezuZU8zqnSdZueMER84UWveHB/owsU8LJvVtSefIQCdGKISwB0lqaiFJjRCeQdM0dp3I5bMdJ/hi16kay0R0bxHEpD4tGd87mmZNrm4wsxDCtUhSUwtJaoTwPKUVJr4/cIaVO07w/YFs65R0L72O6zs3Z3LflozoEi6rmAvhxiSpqYUkNUJ4tvOFZXyRcpLPdp5k94mqwczBft6M7RXFpL4t6RMT0iDTw81mjdzics4VlnGhqIzzhWVcKCxT3xeWcb6ojBA/I/1ah9KvdSiRwb71fk0hGitJamohSY0QjcehrHxW7jjJ6p0nycwrse5vFxbApL4tmNCnBS1D1fRwTdMoLjdxrqAqQbE8qn9/obCc85Xf5xSVYcvM9BYhfvRvE2pNcmIjg2RwsxBXSJKaWkhSI0TjYzJr/Jh2ls92nGTt3kyKy6umh3cIb0JhaQXnC8sorTBf4ip1C/T1ommAUT38jYQGGGkWYCTE38jp3GJ+Tb/Agcy8ixKgAKOBPq2qkpw+rUJkiroQdZCkphaS1AjRuBWUVvDNntN8tuMkyUfOXfS80aCnaUBVYhIaYKSpvzdNA3xoGuCtvq+WwIT4GzF66a/odVMycth+7AK/HjvPzowcCqpVawZV0LBzRKC1Nad/66a0DPWTSspCIElNrSSpEUJYnLhQxOHsAkL9jdZEJsBocEgSYTJrHMzK59djF9hRmegcP1980XHhgT7Wlpx+rUPpFh18RUmUEJ5GkppaSFIjhHBV2XklbD92obI15wK/ncql3FTzV7OPl55eLUPo1yaUvq1CadbEiJdeh0Gvw0uvr/yqvq++7aXXYzBUe06nk9XXhVuRpKYWktQIIdxFSbmJ3Sdy+fXYeXZUJjsXqtXjqS+9jmrJj/53SZAOH28DMU39aRcWQNtqj+gQPxngLBxOkppaSFIjhHBXmqZx5Gwh29NVgrPrRA6FZRWYTBoVZg2zpr5avjeZNSrMZptmaF0Jo5eets0qk5zm6qsl8WkaYJQxQMIuJKmphSQ1QojGxmzWMGkqyVGJTlXCYzJrVJhq7rc8CkoryDhfyJGzhRw5U8jRs4UcO1d4UZdYdUG+XrRt3uSi1p22YQEE+Hg58K6Fp7Hl81v+pwkhhIfS63Xo0eF9FQWVB7VvVuN7k1nj5IVijpwt4OjZQuvjyJlCTuUWk1dSwa7jOew6nnPRtSKCfCoTHJX0RIf44e9jwM/bgL9RPXy91UKn/kYDPl56afURV0VaaoQQQtRLSbmJY+eKOHq2oEbrztGzhZwvLLP5ejod1oTHz6iSHz+jF/7eld8bDb/bVsmQb+V+L4Ou8jqVXyuvqbZ11bar9qvv1Pc6axy6atvqq16nw9fbULnavErEAoxe+PsY8DbI7DR7kJYaIYQQDuPrbaBzZGCtq6TnFJVd1LKTnV9CUZmJ4nITxWUm63ZZZRFETYOiyv3uxGjQ4+9jUEmOsVrS4/O7r0YD/j5e1ZIi9X2A0UCQnzdtmgXI9P2rdFVJzRtvvMGLL75IZmYmvXr14rXXXmPgwIGXPW/58uXcfvvtjB8/ntWrV1v3z5w5k6VLl9Y4NiEhgbVr1150jdLSUuLi4ti1axc7d+6kd+/eV3MLQgghHCDE30ifVkb6tAq97LEVJrNKdH6X7NTcrqixv7jMRNHvtk1mM5Y+CE0DDa3adqXf7a/chaXzQvvdfss3Zk21TBWVmSgsq6Co1ESZSSVjZSYzZUXmGivHXw0vvY4O4U2IjQwkNiqI2MhAukQFER7oI91yl2FzUrNixQrmzp3LokWLiIuL49VXXyUhIYHU1FTCw8PrPC89PZ1HH32UoUOH1vp8YmIiS5YssX7v4+NT63GPP/440dHR7Nq1y9bQhRBCuDAvg55Ag97tlowoqzBTbElyykwUlVVQWFr5tcxEUWm1/ZXfF1Y7rvq5Z/NLyS+t4EBmPgcy8yHllPV1Qv29iY0MoktUELFRgXSJDKJjRBN8r2bQlIeyOal5+eWXmTNnDrNmzQJg0aJFrFmzhnfeeYd58+bVeo7JZGLq1Kk888wzbN68mZycnIuO8fHxITIy8pKv/c0337Bu3TpWrlzJN998Y2voQgghRIMzeukxeukJ9q9/MqZpGqdySzhwOo8Dmfnsr/x65EwBF4rKST5yrsYyH3odtA0LIDYqiK6VrTqxUUFEB/s2ylYdm5KasrIytm/fzvz586379Ho98fHxJCcn13nes88+S3h4OLNnz2bz5s21HpOUlER4eDihoaGMGDGCv//97zRrVjX6Pisrizlz5rB69Wr8/f0vG2tpaSmlpaXW7/Py8q7kFoUQQgin0el0tAjxo0WIHyO7RFj3l5SbOJRVwP7MPA6czudAZh77T+dxoaictDOFpJ0pZM3u09bjA3296BKpWnRiK792jgj0+On1Nt3d2bNnMZlMRERE1NgfERHBgQMHaj1ny5YtvP3226SkpNR53cTERCZNmkTbtm1JS0vjL3/5C6NHjyY5ORmDwYCmacycOZN7772X/v37k56eftlYFy5cyDPPPGPL7QkhhBAuydfbQI+WwfRoGWzdp2kaZ/JL2VfZmmNp3TmcXUB+SQXb0s+zLf289XhdZatO32orxHdo3sSjls2wa8qWn5/PtGnTWLx4MWFhYXUed9ttt1m3e/ToQc+ePWnfvj1JSUmMHDmS1157jfz8/BotRJczf/585s6da/0+Ly+PmJiYq7sRIYQQwsXodDrCg3wJD/Ll+s5VY1rLKsyknSngQGWrjiXpOZNfypEzagbap9tPAKpFp0+rUPq1CqVv6xB6x4S43Zim6mxKasLCwjAYDGRlZdXYn5WVVet4mLS0NNLT0xk7dqx1n9msRol7eXmRmppK+/btLzqvXbt2hIWFcfjwYUaOHMnGjRtJTk6+aPBw//79mTp16kUzp0CN0alrsLEQQgjhqYxeerpEqQHF9Knaf7aglD0ncq2Lp6YczyG/pIIfDp7hh4NnANWa0zkikH6tQ60tOq2b+bvN+Bybi+/FxcUxcOBAXnvtNUAlKa1ateKBBx64aKBwSUkJhw8frrHvb3/7G/n5+fz73/+mU6dOGI3Gi17jxIkTtGrVitWrVzNu3DgyMjJqjIk5deoUCQkJfPrpp8TFxdGyZcvLxi3F94QQQogqFSYzBzLz2ZGhkpwdGRc4fr74ouOaBRjpWy3J6dky2KEzruxafG/u3LnMmDGD/v37M3DgQF599VUKCwuts6GmT59OixYtWLhwIb6+vnTv3r3G+SEhIQDW/QUFBTzzzDNMnjyZyMhI0tLSePzxx+nQoQMJCQkAtGrVqsY1mjRpAkD79u2vKKERQgghRE1eBj3dWwTTvUUw0we1ASA7r6RakpPDnhO5nCss47t9WXy3T/XSeOl1dGsRTN9WIdaxOVHBfk68kyo2JzVTpkzhzJkzPPXUU2RmZtK7d2/Wrl1rHTyckZGBXn/llRANBgO7d+9m6dKl5OTkEB0dzahRo3juueek+0gIIYRwoPAgXxK7R5HYPQqA0goTe0/msbMy0fn12AXO5Jda1/lasjUdgKhgX/q2VmNzZg5u47TBx7L2kxBCCCGuiKZpnLhQzI6MC+w4doHtGRfYfzofk1mlEq2a+vPD48Mb9DVl7SchhBBCNDidTkdMU39imvozvncLAIrKKth1PJcdGRfwcfKaVZLUCCGEEOKq+Ru9GNS+GYPaN7v8wXYmy4AKIYQQwiNIUiOEEEIIjyBJjRBCCCE8giQ1QgghhPAIktQIIYQQwiNIUiOEEEIIjyBJjRBCCCE8giQ1QgghhPAIktQIIYQQwiNIUiOEEEIIjyBJjRBCCCE8giQ1QgghhPAIktQIIYQQwiM0mlW6NU0DIC8vz8mRCCGEEOJKWT63LZ/jl9Jokpr8/HwAYmJinByJEEIIIWyVn59PcHDwJY/RaVeS+ngAs9nMqVOnCAwMRKfTNei18/LyiImJ4fjx4wQFBTXotV1NY7pXaFz3K/fquRrT/cq9eh5N08jPzyc6Ohq9/tKjZhpNS41er6dly5Z2fY2goCCP/o9VXWO6V2hc9yv36rka0/3KvXqWy7XQWMhAYSGEEEJ4BElqhBBCCOERJKlpAD4+PixYsAAfHx9nh2J3jeleoXHdr9yr52pM9yv32rg1moHCQgghhPBs0lIjhBBCCI8gSY0QQgghPIIkNUIIIYTwCJLUCCGEEMIjSFJzhd544w3atGmDr68vcXFxbNu27ZLHf/LJJ8TGxuLr60uPHj34+uuvHRTp1Vu4cCEDBgwgMDCQ8PBwJkyYQGpq6iXPeffdd9HpdDUevr6+Doq4fp5++umLYo+Njb3kOe74vgK0adPmonvV6XTcf//9tR7vTu/rDz/8wNixY4mOjkan07F69eoaz2uaxlNPPUVUVBR+fn7Ex8dz6NChy17X1p95R7nU/ZaXl/PEE0/Qo0cPAgICiI6OZvr06Zw6deqS17yanwVHuNx7O3PmzIviTkxMvOx1XfG9vdy91vbzq9PpePHFF+u8pqu+r/YkSc0VWLFiBXPnzmXBggXs2LGDXr16kZCQQHZ2dq3H//jjj9x+++3Mnj2bnTt3MmHCBCZMmMDevXsdHLltNm3axP33389PP/3Ed999R3l5OaNGjaKwsPCS5wUFBXH69Gnr49ixYw6KuP66detWI/YtW7bUeay7vq8Av/zyS437/O677wC45ZZb6jzHXd7XwsJCevXqxRtvvFHr8//85z/5v//7PxYtWsTPP/9MQEAACQkJlJSU1HlNW3/mHelS91tUVMSOHTt48skn2bFjB5999hmpqamMGzfuste15WfBUS733gIkJibWiPujjz665DVd9b293L1Wv8fTp0/zzjvvoNPpmDx58iWv64rvq11p4rIGDhyo3X///dbvTSaTFh0drS1cuLDW42+99VbtpptuqrEvLi5Ou+eee+waZ0PLzs7WAG3Tpk11HrNkyRItODjYcUE1oAULFmi9evW64uM95X3VNE3705/+pLVv314zm821Pu+u7yugrVq1yvq92WzWIiMjtRdffNG6LycnR/Px8dE++uijOq9j68+8s/z+fmuzbds2DdCOHTtW5zG2/iw4Q233OmPGDG38+PE2Xccd3tsreV/Hjx+vjRgx4pLHuMP72tCkpeYyysrK2L59O/Hx8dZ9er2e+Ph4kpOTaz0nOTm5xvEACQkJdR7vqnJzcwFo2rTpJY8rKCigdevWxMTEMH78eH777TdHhNcgDh06RHR0NO3atWPq1KlkZGTUeaynvK9lZWW8//773HnnnZdc3NWd31eLo0ePkpmZWeN9Cw4OJi4urs737Wp+5l1Zbm4uOp2OkJCQSx5ny8+CK0lKSiI8PJzOnTtz3333ce7cuTqP9ZT3NisrizVr1jB79uzLHuuu7+vVkqTmMs6ePYvJZCIiIqLG/oiICDIzM2s9JzMz06bjXZHZbObhhx/m2muvpXv37nUe17lzZ9555x0+//xz3n//fcxmM4MHD+bEiRMOjPbqxMXF8e6777J27VrefPNNjh49ytChQ8nPz6/1eE94XwFWr15NTk4OM2fOrPMYd35fq7O8N7a8b1fzM++qSkpKeOKJJ7j99tsvueChrT8LriIxMZFly5axYcMGXnjhBTZt2sTo0aMxmUy1Hu8p7+3SpUsJDAxk0qRJlzzOXd/X+mg0q3QL29x///3s3bv3sv2vgwYNYtCgQdbvBw8eTJcuXXjrrbd47rnn7B1mvYwePdq63bNnT+Li4mjdujUff/zxFf0F5K7efvttRo8eTXR0dJ3HuPP7KpTy8nJuvfVWNE3jzTffvOSx7vqzcNttt1m3e/ToQc+ePWnfvj1JSUmMHDnSiZHZ1zvvvMPUqVMvO3jfXd/X+pCWmssICwvDYDCQlZVVY39WVhaRkZG1nhMZGWnT8a7mgQce4KuvvuL777+nZcuWNp3r7e1Nnz59OHz4sJ2is5+QkBA6depUZ+zu/r4CHDt2jPXr13PXXXfZdJ67vq+W98aW9+1qfuZdjSWhOXbsGN99990lW2lqc7mfBVfVrl07wsLC6ozbE97bzZs3k5qaavPPMLjv+2oLSWouw2g00q9fPzZs2GDdZzab2bBhQ42/ZKsbNGhQjeMBvvvuuzqPdxWapvHAAw+watUqNm7cSNu2bW2+hslkYs+ePURFRdkhQvsqKCggLS2tztjd9X2tbsmSJYSHh3PTTTfZdJ67vq9t27YlMjKyxvuWl5fHzz//XOf7djU/867EktAcOnSI9evX06xZM5uvcbmfBVd14sQJzp07V2fc7v7egmpp7devH7169bL5XHd9X23i7JHK7mD58uWaj4+P9u6772r79u3T7r77bi0kJETLzMzUNE3Tpk2bps2bN896/NatWzUvLy/tpZde0vbv368tWLBA8/b21vbs2eOsW7gi9913nxYcHKwlJSVpp0+ftj6Kioqsx/z+Xp955hnt22+/1dLS0rTt27drt912m+br66v99ttvzrgFm/z5z3/WkpKStKNHj2pbt27V4uPjtbCwMC07O1vTNM95Xy1MJpPWqlUr7YknnrjoOXd+X/Pz87WdO3dqO3fu1ADt5Zdf1nbu3Gmd7fP8889rISEh2ueff67t3r1bGz9+vNa2bVutuLjYeo0RI0Zor732mvX7y/3MO9Ol7resrEwbN26c1rJlSy0lJaXGz3Fpaan1Gr+/38v9LDjLpe41Pz9fe/TRR7Xk5GTt6NGj2vr167W+fftqHTt21EpKSqzXcJf39nL/jzVN03JzczV/f3/tzTffrPUa7vK+2pMkNVfotdde01q1aqUZjUZt4MCB2k8//WR97rrrrtNmzJhR4/iPP/5Y69Spk2Y0GrVu3bppa9ascXDEtgNqfSxZssR6zO/v9eGHH7b+u0RERGg33nijtmPHDscHfxWmTJmiRUVFaUajUWvRooU2ZcoU7fDhw9bnPeV9tfj22281QEtNTb3oOXd+X7///vta/99a7sdsNmtPPvmkFhERofn4+GgjR4686N+gdevW2oIFC2rsu9TPvDNd6n6PHj1a58/x999/b73G7+/3cj8LznKpey0qKtJGjRqlNW/eXPP29tZat26tzZkz56LkxF3e28v9P9Y0TXvrrbc0Pz8/LScnp9ZruMv7ak86TdM0uzYFCSGEEEI4gIypEUIIIYRHkKRGCCGEEB5BkhohhBBCeARJaoQQQgjhESSpEUIIIYRHkKRGCCGEEB5BkhohhBBCeARJaoQQQgjhESSpEUIIIYRHkKRGCCGEEB5BkhohhBBCeARJaoQQQgjhEf4/vZJzXm3xhtIAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "this_history_df[['loss', 'val_loss']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "2342a31f-4891-4834-851a-b90b8b729c3c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "this_history_df[['auc', 'val_auc']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "cee80637-cabf-4974-b476-08c5fba165ff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "this_history_df[['auc_roc', 'val_auc_roc']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "e184b0ca-1d31-4798-88a0-4e590b67062e",
   "metadata": {},
   "outputs": [],
   "source": [
    "this_model.save(Path(setup.temp_path).joinpath('23_DATA_model.keras'),save_format='tf')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d326c9ea-4759-4418-a64d-56c151049c25",
   "metadata": {},
   "source": [
    "## Predictions on all data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "cd6910ee-5ffe-4def-9aa8-7dba76446b08",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  117/21503 [..............................] - ETA: 27s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-03-29 00:43:05.070817: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "21503/21503 [==============================] - 26s 1ms/step\n"
     ]
    }
   ],
   "source": [
    "all_predictions = this_model.predict(X.drop(columns='dset'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "98a2cce1-b7d3-4ebc-8869-80705297e145",
   "metadata": {},
   "outputs": [],
   "source": [
    "all_predictions_df = pd.DataFrame(all_predictions, index=X.index) \\\n",
    "    .set_axis(['predict_prob'], axis=1) \\\n",
    "    .reset_index() \\\n",
    "    .merge(business_data[['target', 'LoanNr_ChkDgt', 'dset', 'dset_naics_holdout']], on='LoanNr_ChkDgt')\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "3fc7da9c-6bd2-4d1b-b77f-b62679825d5d",
   "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>predict_prob</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>predict_prob</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.315101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <td>0.315101</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              predict_prob    target\n",
       "predict_prob      1.000000  0.315101\n",
       "target            0.315101  1.000000"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_predictions_df[['predict_prob', 'target']].corr(method='spearman')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e79afc4e-cb67-4064-9b19-dc52500baef2",
   "metadata": {},
   "source": [
    "##### Threshold Tune & Binary Predictions\n",
    "Using training probability predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "ccd0bc23-fbc0-40d0-8d8b-f2302405a875",
   "metadata": {},
   "outputs": [],
   "source": [
    "all_pred_train = all_predictions_df[all_predictions_df['dset'] == 'train']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "df36d9e9-f35e-46b3-a5be-08081f998319",
   "metadata": {},
   "outputs": [],
   "source": [
    "thresh_tune_data = sg_plot.get_f1_frame(all_pred_train['target'], \n",
    "                                        all_pred_train['predict_prob'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "de634156-caf6-4a52-b5d7-44b8e986aa1d",
   "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>thresh</th>\n",
       "      <th>f1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.18</td>\n",
       "      <td>0.448898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.20</td>\n",
       "      <td>0.448376</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.16</td>\n",
       "      <td>0.445598</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    thresh        f1\n",
       "9     0.18  0.448898\n",
       "10    0.20  0.448376\n",
       "8     0.16  0.445598"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "thresh_tune_data.sort_values('f1', ascending=False, inplace=True)\n",
    "thresh_tune_data.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "c746b880-d781-4bc1-9df2-89211127c22a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.18"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_thresh = thresh_tune_data['thresh'].iloc[0]\n",
    "best_thresh"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c0ac5e2-be0d-42f0-960d-1f81b4a5d4e2",
   "metadata": {},
   "source": [
    "##### Append binary predictions to probability predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "641da72e-1bca-4fe7-96c4-6c029b521bd7",
   "metadata": {},
   "outputs": [],
   "source": [
    "all_predictions_df['predict_bin'] = sg_plot.get_binary_predictions(all_predictions_df['predict_prob'], best_thresh)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "9f6ecdf4-ac01-4c39-8315-e8583daa337e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "predict_bin\n",
       "0    0.541542\n",
       "1    0.458458\n",
       "Name: proportion, dtype: float64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_predictions_df['predict_bin'].value_counts(normalize=True, dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "cbfed830-7ff8-40f6-a691-6b894b26d6bc",
   "metadata": {},
   "outputs": [],
   "source": [
    "all_predictions_df.to_parquet(Path(setup.temp_path).joinpath('23_DATA_predictions.parquet'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69903966-6baa-4579-8e84-ce38e8ebdd77",
   "metadata": {},
   "source": [
    "## Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "09b7437b-4b51-42c4-a267-fb244663df49",
   "metadata": {},
   "outputs": [],
   "source": [
    "all_predictions_df = pd.read_parquet(Path(setup.temp_path).joinpath('23_DATA_predictions.parquet'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "0c748e1c-ddb3-4d92-8f84-485736633e10",
   "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>0.629367</td>\n",
       "      <td>0.435506</td>\n",
       "      <td>0.310319</td>\n",
       "      <td>0.729996</td>\n",
       "      <td>0.362163</td>\n",
       "      <td>0.724914</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>train</td>\n",
       "      <td>0.634990</td>\n",
       "      <td>0.448898</td>\n",
       "      <td>0.324815</td>\n",
       "      <td>0.726388</td>\n",
       "      <td>0.380671</td>\n",
       "      <td>0.727427</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>val</td>\n",
       "      <td>0.632653</td>\n",
       "      <td>0.448230</td>\n",
       "      <td>0.325581</td>\n",
       "      <td>0.719134</td>\n",
       "      <td>0.376875</td>\n",
       "      <td>0.722401</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>test</td>\n",
       "      <td>0.632690</td>\n",
       "      <td>0.447950</td>\n",
       "      <td>0.324200</td>\n",
       "      <td>0.724497</td>\n",
       "      <td>0.376685</td>\n",
       "      <td>0.724844</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>test</td>\n",
       "      <td>0.624593</td>\n",
       "      <td>0.417035</td>\n",
       "      <td>0.290489</td>\n",
       "      <td>0.738939</td>\n",
       "      <td>0.341251</td>\n",
       "      <td>0.725387</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        0.629367  0.435506         0.310319      0.729996   \n",
       "1  train        0.634990  0.448898         0.324815      0.726388   \n",
       "2    val        0.632653  0.448230         0.325581      0.719134   \n",
       "0   test        0.632690  0.447950         0.324200      0.724497   \n",
       "1   test        0.624593  0.417035         0.290489      0.738939   \n",
       "\n",
       "   average_precision_score  roc_auc_score  dset_naics_holdout  \n",
       "0                 0.362163       0.724914                 NaN  \n",
       "1                 0.380671       0.727427                 NaN  \n",
       "2                 0.376875       0.722401                 NaN  \n",
       "0                 0.376685       0.724844                 0.0  \n",
       "1                 0.341251       0.725387                 1.0  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics_dset_df = all_predictions_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 = all_predictions_df[all_predictions_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('23_REPORT_metrics.csv'), index=True)\n",
    "metrics_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2bf6b2f6-992d-4c4b-88cb-02eb418ae751",
   "metadata": {},
   "source": [
    "## Embeddings\n",
    "Save the hidden layer weight prior to sigmoid output, for possible later use in unsupervised GNN "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "f8bf0603-5235-4344-a73d-a5d632befd0d",
   "metadata": {},
   "outputs": [],
   "source": [
    "embed_model = Model(inputs=this_model.inputs, outputs=this_model.layers[-2].output) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "0ee68a3c-a630-4fde-b765-f441d2bb3895",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  187/21503 [..............................] - ETA: 17s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-03-29 00:43:51.171577: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "21503/21503 [==============================] - 17s 802us/step\n"
     ]
    }
   ],
   "source": [
    "embed_out = embed_model.predict(X.drop(columns='dset'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "f622987d-4206-494f-8482-08ac99fff2f9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(688081, 64)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embed_out.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "2193800a-3e35-460c-8ee0-576d460f53bb",
   "metadata": {},
   "outputs": [],
   "source": [
    "embed_df = pd.DataFrame(embed_out, index=X.index) \n",
    "embed_df.columns = [f'emb_{i:03d}' for i in range(len(embed_df.columns))]\n",
    "embed_df.reset_index(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "17f1e84b-2aec-4b6b-bd27-04a79008b2de",
   "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>LoanNr_ChkDgt</th>\n",
       "      <th>emb_000</th>\n",
       "      <th>emb_001</th>\n",
       "      <th>emb_002</th>\n",
       "      <th>emb_003</th>\n",
       "      <th>emb_004</th>\n",
       "      <th>emb_005</th>\n",
       "      <th>emb_006</th>\n",
       "      <th>emb_007</th>\n",
       "      <th>emb_008</th>\n",
       "      <th>...</th>\n",
       "      <th>emb_054</th>\n",
       "      <th>emb_055</th>\n",
       "      <th>emb_056</th>\n",
       "      <th>emb_057</th>\n",
       "      <th>emb_058</th>\n",
       "      <th>emb_059</th>\n",
       "      <th>emb_060</th>\n",
       "      <th>emb_061</th>\n",
       "      <th>emb_062</th>\n",
       "      <th>emb_063</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1000014003</td>\n",
       "      <td>0.999997</td>\n",
       "      <td>-0.185493</td>\n",
       "      <td>0.778988</td>\n",
       "      <td>0.999991</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.058228</td>\n",
       "      <td>-0.999978</td>\n",
       "      <td>-0.427396</td>\n",
       "      <td>0.351970</td>\n",
       "      <td>...</td>\n",
       "      <td>0.577731</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-0.999999</td>\n",
       "      <td>0.883864</td>\n",
       "      <td>0.855696</td>\n",
       "      <td>-0.352030</td>\n",
       "      <td>-0.999815</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999773</td>\n",
       "      <td>0.999999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1000024006</td>\n",
       "      <td>0.999988</td>\n",
       "      <td>0.156657</td>\n",
       "      <td>0.977533</td>\n",
       "      <td>0.999999</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.979082</td>\n",
       "      <td>-0.999991</td>\n",
       "      <td>-0.969789</td>\n",
       "      <td>-0.513925</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.656893</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-0.999985</td>\n",
       "      <td>0.998748</td>\n",
       "      <td>0.955726</td>\n",
       "      <td>0.141398</td>\n",
       "      <td>-0.999807</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999900</td>\n",
       "      <td>0.999999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1000034009</td>\n",
       "      <td>0.999998</td>\n",
       "      <td>-0.455601</td>\n",
       "      <td>0.968171</td>\n",
       "      <td>0.475721</td>\n",
       "      <td>-0.999759</td>\n",
       "      <td>0.998743</td>\n",
       "      <td>-0.999965</td>\n",
       "      <td>-0.916335</td>\n",
       "      <td>0.151704</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.590935</td>\n",
       "      <td>-0.999942</td>\n",
       "      <td>-0.999999</td>\n",
       "      <td>0.999960</td>\n",
       "      <td>0.952065</td>\n",
       "      <td>-0.222828</td>\n",
       "      <td>-0.999999</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999772</td>\n",
       "      <td>0.615164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1000084002</td>\n",
       "      <td>0.999999</td>\n",
       "      <td>-0.584501</td>\n",
       "      <td>0.894935</td>\n",
       "      <td>0.050126</td>\n",
       "      <td>-0.999244</td>\n",
       "      <td>0.993820</td>\n",
       "      <td>-0.999929</td>\n",
       "      <td>-0.585380</td>\n",
       "      <td>0.448615</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.165131</td>\n",
       "      <td>-0.999978</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.999703</td>\n",
       "      <td>0.945566</td>\n",
       "      <td>-0.459805</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999757</td>\n",
       "      <td>0.447167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1000094005</td>\n",
       "      <td>0.999978</td>\n",
       "      <td>0.155097</td>\n",
       "      <td>0.991113</td>\n",
       "      <td>0.999997</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.988084</td>\n",
       "      <td>-0.999998</td>\n",
       "      <td>-0.978370</td>\n",
       "      <td>-0.442554</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.695188</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-0.999988</td>\n",
       "      <td>0.999535</td>\n",
       "      <td>0.958219</td>\n",
       "      <td>0.319526</td>\n",
       "      <td>-0.999871</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999939</td>\n",
       "      <td>0.999997</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 65 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  LoanNr_ChkDgt   emb_000   emb_001   emb_002   emb_003   emb_004   emb_005  \\\n",
       "0    1000014003  0.999997 -0.185493  0.778988  0.999991 -1.000000  0.058228   \n",
       "1    1000024006  0.999988  0.156657  0.977533  0.999999 -1.000000  0.979082   \n",
       "2    1000034009  0.999998 -0.455601  0.968171  0.475721 -0.999759  0.998743   \n",
       "3    1000084002  0.999999 -0.584501  0.894935  0.050126 -0.999244  0.993820   \n",
       "4    1000094005  0.999978  0.155097  0.991113  0.999997 -1.000000  0.988084   \n",
       "\n",
       "    emb_006   emb_007   emb_008  ...   emb_054   emb_055   emb_056   emb_057  \\\n",
       "0 -0.999978 -0.427396  0.351970  ...  0.577731 -1.000000 -0.999999  0.883864   \n",
       "1 -0.999991 -0.969789 -0.513925  ... -0.656893 -1.000000 -0.999985  0.998748   \n",
       "2 -0.999965 -0.916335  0.151704  ... -0.590935 -0.999942 -0.999999  0.999960   \n",
       "3 -0.999929 -0.585380  0.448615  ... -0.165131 -0.999978 -1.000000  0.999703   \n",
       "4 -0.999998 -0.978370 -0.442554  ... -0.695188 -1.000000 -0.999988  0.999535   \n",
       "\n",
       "    emb_058   emb_059   emb_060  emb_061   emb_062   emb_063  \n",
       "0  0.855696 -0.352030 -0.999815      1.0  0.999773  0.999999  \n",
       "1  0.955726  0.141398 -0.999807      1.0  0.999900  0.999999  \n",
       "2  0.952065 -0.222828 -0.999999      1.0  0.999772  0.615164  \n",
       "3  0.945566 -0.459805 -1.000000      1.0  0.999757  0.447167  \n",
       "4  0.958219  0.319526 -0.999871      1.0  0.999939  0.999997  \n",
       "\n",
       "[5 rows x 65 columns]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embed_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "6a8d3c99-24d1-404c-a867-b67f6dd6a4a9",
   "metadata": {},
   "outputs": [],
   "source": [
    "embed_df.to_parquet(Path(setup.temp_path).joinpath('23_DATA_embeddings.parquet'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "61477278-45ce-4742-b352-05fdde43506f",
   "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
}