23_downsample_hier.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": [
    "## NN, Downsample train, NAICS only embedding\n",
    "Do the embedding model again but downsample training.  This is a baseline for comparing random vs. the custom data generator"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e169c2e8-ba0f-488f-acf6-87d419bd4e99",
   "metadata": {},
   "source": [
    "Use the dataset with no missing values. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d47a7ec9-7d7b-47df-a902-b2f1e54f1135",
   "metadata": {},
   "source": [
    "*This script takes about 15 minutes 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\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "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",
    "import keras\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": "9f1e4f2e-0e45-4e28-8e9c-212e0bb51199",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sba_nn.sba_nn import sbnn_metrics, sbnn_model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3cbc00d2-bc83-42d9-83eb-65fcf306dccd",
   "metadata": {},
   "source": [
    "## Input Data, Downsample Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e014cd31-0f8d-42e3-82f9-b529eec8a3cc",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "sba_loans = pd.read_parquet(Path(setup.temp_path).joinpath('20_DATA_downsamp_train.parquet'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "45864963-90d7-43e7-89eb-03e947feec17",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dset\n",
       "test     153048\n",
       "val       95604\n",
       "train     44684\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sba_loans['dset'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "8e65cb7c-792c-46ff-a530-0d5c64240d60",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(Path(setup.temp_path).joinpath('01_DATA_features.pkl'), 'rb') as fin:\n",
    "    imputer_features = pickle.load(fin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "6f0ce28e-a80c-415b-b068-5c4d49370da8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'NAICS': 1170,\n",
       " 'NAICS_5': 764,\n",
       " 'NAICS_4': 345,\n",
       " 'NAICS_3': 107,\n",
       " 'NAICS_sector': 21}"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with open(Path(setup.temp_path).joinpath('01_DATA_naics_max_encodings.pkl'), 'rb') as fin:\n",
    "    naics_max_levels= pickle.load(fin)\n",
    "naics_max_levels"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4d4116da-d689-478a-9c5d-a330f231e615",
   "metadata": {},
   "source": [
    "##### Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f98c6e41-38d0-4d5f-b821-e3837b8c2068",
   "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']\n"
     ]
    }
   ],
   "source": [
    "# Numeric features\n",
    "features_numeric = [f for f in imputer_features if 'NAICS' not in f]\n",
    "features_numeric_len = len(features_numeric)\n",
    "print(features_numeric)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "a74fa386-4011-45ee-bdb5-d671db9ffa91",
   "metadata": {},
   "outputs": [],
   "source": [
    "# NAICS feature \n",
    "features_naics = ['NAICS', 'NAICS_4', 'NAICS_3', 'NAICS_sector']\n",
    "features_naics_max_levels  = [naics_max_levels[n] for n in features_naics]\n",
    "features_naics_emb_dim = [setup.nn_naics_embed_size_dict[n] for n in features_naics]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "867031ee-8b6e-424d-b497-8e19409ba1d9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1170, 345, 107, 21]\n",
      "[8, 4, 4, 2]\n"
     ]
    }
   ],
   "source": [
    "print(features_naics_max_levels)\n",
    "print(features_naics_emb_dim)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b979396c-c822-4940-a292-6e07f1b54f58",
   "metadata": {},
   "source": [
    "##### Datasets for train, validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "43bca011-a59d-4496-9f03-87f4e350a253",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = sba_loans[['dset', 'LoanNr_ChkDgt'] + features_numeric].set_index('LoanNr_ChkDgt').sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "aa162d01-ed73-4ae3-832e-4aeea09b70f4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training X: (44684, 10), y:(44684,)\n"
     ]
    }
   ],
   "source": [
    "X_train = X[X['dset'] == 'train'].drop(columns='dset')\n",
    "y_train = sba_loans[sba_loans['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": 15,
   "id": "b32b5fd3-6f6e-4aec-8e20-0e2b71a4814c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "val X: (95604, 10), y:(95604,)\n"
     ]
    }
   ],
   "source": [
    "X_val = X[X['dset'] == 'val'].drop(columns='dset')\n",
    "y_val = sba_loans[sba_loans['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": 16,
   "id": "01b8ff96-159b-4283-b5fb-c4b5dbd277d5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.20295855339718916\n"
     ]
    }
   ],
   "source": [
    "base_thresh = y_train.mean()\n",
    "print(base_thresh)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "1b8ce4c9-6907-4b70-982a-b92a52b3159d",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_naics = [sba_loans[['dset', 'LoanNr_ChkDgt', n]].set_index('LoanNr_ChkDgt').sort_index() \\\n",
    "           for n in features_naics]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "a6f129e0-a4d0-4d93-9886-153c9803d403",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_naics_train = [Xn[Xn['dset'] == 'train'].drop(columns='dset') for Xn in X_naics]\n",
    "X_naics_val =  [Xn[Xn['dset'] == 'val'].drop(columns='dset') for Xn in X_naics]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3db4ea1c-77b2-4721-b106-1e66fc502139",
   "metadata": {},
   "source": [
    "## Create, fit model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "4d48ddfa-580c-4321-9917-0bd176b67d87",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-05-20 13:47:54.537691: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M1\n",
      "2024-05-20 13:47:54.537748: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 16.00 GB\n",
      "2024-05-20 13:47:54.537763: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 5.33 GB\n",
      "2024-05-20 13:47:54.537822: 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",
      "2024-05-20 13:47:54.537842: 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 = sbnn_model.create_emb_model(features_numeric_len,\n",
    "                                         features_naics_max_levels,\n",
    "                                         features_naics_emb_dim,\n",
    "                                         naics_embedding_names = features_naics,\n",
    "                                         hidden_size = setup.nn_layer_sizes,\n",
    "                                         activation='tanh',\n",
    "                                         lr=setup.nn_learning_rate,\n",
    "                                         opt_func = setup.nn_optimizer,\n",
    "                                         dropout = setup.nn_dropout)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "47036621-ef13-4957-997b-f8ef86016f7f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                Output Shape                 Param #   Connected to                  \n",
      "==================================================================================================\n",
      " input_naics_00 (InputLayer  [(None, 1)]                  0         []                            \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " input_naics_01 (InputLayer  [(None, 1)]                  0         []                            \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " input_naics_02 (InputLayer  [(None, 1)]                  0         []                            \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " input_naics_03 (InputLayer  [(None, 1)]                  0         []                            \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " emb_naics_00 (Embedding)    (None, 1, 8)                 9360      ['input_naics_00[0][0]']      \n",
      "                                                                                                  \n",
      " emb_naics_01 (Embedding)    (None, 1, 4)                 1380      ['input_naics_01[0][0]']      \n",
      "                                                                                                  \n",
      " emb_naics_02 (Embedding)    (None, 1, 4)                 428       ['input_naics_02[0][0]']      \n",
      "                                                                                                  \n",
      " emb_naics_03 (Embedding)    (None, 1, 2)                 42        ['input_naics_03[0][0]']      \n",
      "                                                                                                  \n",
      " NAICS (Flatten)             (None, 8)                    0         ['emb_naics_00[0][0]']        \n",
      "                                                                                                  \n",
      " NAICS_4 (Flatten)           (None, 4)                    0         ['emb_naics_01[0][0]']        \n",
      "                                                                                                  \n",
      " NAICS_3 (Flatten)           (None, 4)                    0         ['emb_naics_02[0][0]']        \n",
      "                                                                                                  \n",
      " NAICS_sector (Flatten)      (None, 2)                    0         ['emb_naics_03[0][0]']        \n",
      "                                                                                                  \n",
      " input_numeric (InputLayer)  [(None, 10)]                 0         []                            \n",
      "                                                                                                  \n",
      " input_concat (Concatenate)  (None, 28)                   0         ['NAICS[0][0]',               \n",
      "                                                                     'NAICS_4[0][0]',             \n",
      "                                                                     'NAICS_3[0][0]',             \n",
      "                                                                     'NAICS_sector[0][0]',        \n",
      "                                                                     'input_numeric[0][0]']       \n",
      "                                                                                                  \n",
      " layer_00 (Dense)            (None, 128)                  3712      ['input_concat[0][0]']        \n",
      "                                                                                                  \n",
      " dropout_00 (Dropout)        (None, 128)                  0         ['layer_00[0][0]']            \n",
      "                                                                                                  \n",
      " layer_01 (Dense)            (None, 64)                   8256      ['dropout_00[0][0]']          \n",
      "                                                                                                  \n",
      " dropout_01 (Dropout)        (None, 64)                   0         ['layer_01[0][0]']            \n",
      "                                                                                                  \n",
      " output (Dense)              (None, 1)                    65        ['dropout_01[0][0]']          \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 23243 (90.79 KB)\n",
      "Trainable params: 23243 (90.79 KB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "this_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "5b21b5ef-7f75-4708-b8d7-59825d38f601",
   "metadata": {},
   "outputs": [],
   "source": [
    "callback = keras.callbacks.EarlyStopping(patience=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "31772016-3643-468b-8d65-c456168cad6e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-05-20 13:47:56.000263: 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": [
      "1397/1397 [==============================] - ETA: 0s - loss: 0.4830 - auc: 0.2919 - auc_1: 0.6574"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-05-20 13:48:40.483158: 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": [
      "1397/1397 [==============================] - 80s 55ms/step - loss: 0.4830 - auc: 0.2919 - auc_1: 0.6574 - val_loss: 0.4628 - val_auc: 0.3412 - val_auc_1: 0.7048\n",
      "Epoch 2/20\n",
      "1397/1397 [==============================] - 78s 56ms/step - loss: 0.4646 - auc: 0.3353 - auc_1: 0.6990 - val_loss: 0.4607 - val_auc: 0.3491 - val_auc_1: 0.7081\n",
      "Epoch 3/20\n",
      "1397/1397 [==============================] - 84s 60ms/step - loss: 0.4591 - auc: 0.3481 - auc_1: 0.7104 - val_loss: 0.4661 - val_auc: 0.3501 - val_auc_1: 0.7066\n",
      "Epoch 4/20\n",
      "1397/1397 [==============================] - 85s 61ms/step - loss: 0.4566 - auc: 0.3556 - auc_1: 0.7151 - val_loss: 0.4626 - val_auc: 0.3501 - val_auc_1: 0.7078\n",
      "Epoch 5/20\n",
      "1397/1397 [==============================] - 82s 59ms/step - loss: 0.4537 - auc: 0.3634 - auc_1: 0.7203 - val_loss: 0.4629 - val_auc: 0.3517 - val_auc_1: 0.7082\n"
     ]
    }
   ],
   "source": [
    "this_history = this_model.fit([X_train] + X_naics_train, y_train,\n",
    "                              validation_data=([X_val] + X_naics_val, y_val),\n",
    "                              batch_size=setup.nn_batch_size,\n",
    "                              epochs=setup.nn_epochs,\n",
    "                              callbacks=[callback])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "460c4b43-c673-4ee1-b070-d2d596ac4561",
   "metadata": {},
   "outputs": [],
   "source": [
    "this_history_df = sbnn_model.process_history(this_history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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": 25,
   "id": "73e89efd-75a9-4b27-81a8-13e12d5482be",
   "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[['loss', 'val_loss']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "2342a31f-4891-4834-851a-b90b8b729c3c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 26,
     "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": 27,
   "id": "cee80637-cabf-4974-b476-08c5fba165ff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 27,
     "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": 28,
   "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": 29,
   "id": "d5c39a5a-dc19-4606-b744-34865bc6c5c5",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_naics = [sba_loans[['dset', 'LoanNr_ChkDgt', n]].set_index('LoanNr_ChkDgt').sort_index() \\\n",
    "           for n in features_naics]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "cd6910ee-5ffe-4def-9aa8-7dba76446b08",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  13/9167 [..............................] - ETA: 40s  "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-05-20 13:54:47.886977: 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": [
      "9167/9167 [==============================] - 51s 6ms/step\n"
     ]
    }
   ],
   "source": [
    "all_predictions = this_model.predict([X.drop(columns='dset')] +\n",
    "                                     [Xn.drop(columns='dset') for Xn in X_naics])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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(sba_loans[['target', 'LoanNr_ChkDgt', 'dset', 'dset_naics_holdout', 'NAICS']], \n",
    "           on='LoanNr_ChkDgt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>0.296084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <td>0.296084</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              predict_prob    target\n",
       "predict_prob      1.000000  0.296084\n",
       "target            0.296084  1.000000"
      ]
     },
     "execution_count": 32,
     "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": 33,
   "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": 34,
   "id": "df36d9e9-f35e-46b3-a5be-08081f998319",
   "metadata": {},
   "outputs": [],
   "source": [
    "thresh_tune_data = sbnn_metrics.get_f1_frame(all_pred_train['target'], \n",
    "                                        all_pred_train['predict_prob'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>0.455452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.20</td>\n",
       "      <td>0.455395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.16</td>\n",
       "      <td>0.450364</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    thresh        f1\n",
       "9     0.18  0.455452\n",
       "10    0.20  0.455395\n",
       "8     0.16  0.450364"
      ]
     },
     "execution_count": 35,
     "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": 36,
   "id": "c746b880-d781-4bc1-9df2-89211127c22a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.18"
      ]
     },
     "execution_count": 36,
     "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": 37,
   "id": "641da72e-1bca-4fe7-96c4-6c029b521bd7",
   "metadata": {},
   "outputs": [],
   "source": [
    "all_predictions_df['predict_bin'] = sbnn_metrics.get_binary_predictions(all_predictions_df['predict_prob'], best_thresh)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "9f6ecdf4-ac01-4c39-8315-e8583daa337e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "predict_bin\n",
       "0    0.524337\n",
       "1    0.475663\n",
       "Name: proportion, dtype: float64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_predictions_df['predict_bin'].value_counts(normalize=True, dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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": "eb0ecbd2-abff-46d8-8c70-f7bccd363b69",
   "metadata": {},
   "source": [
    "## Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "be4ceefe-c7a8-45bd-8573-2daefe99a1c2",
   "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>0.613605</td>\n",
       "      <td>0.437802</td>\n",
       "      <td>0.309835</td>\n",
       "      <td>0.745854</td>\n",
       "      <td>0.346552</td>\n",
       "      <td>0.709198</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>train</td>\n",
       "      <td>0.636850</td>\n",
       "      <td>0.455452</td>\n",
       "      <td>0.327352</td>\n",
       "      <td>0.748263</td>\n",
       "      <td>0.381655</td>\n",
       "      <td>0.734007</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>val</td>\n",
       "      <td>0.622809</td>\n",
       "      <td>0.436274</td>\n",
       "      <td>0.313700</td>\n",
       "      <td>0.716067</td>\n",
       "      <td>0.351811</td>\n",
       "      <td>0.708278</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>test</td>\n",
       "      <td>0.624088</td>\n",
       "      <td>0.434716</td>\n",
       "      <td>0.312984</td>\n",
       "      <td>0.711409</td>\n",
       "      <td>0.346789</td>\n",
       "      <td>0.704898</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>test</td>\n",
       "      <td>0.596180</td>\n",
       "      <td>0.442513</td>\n",
       "      <td>0.305231</td>\n",
       "      <td>0.804224</td>\n",
       "      <td>0.348779</td>\n",
       "      <td>0.719503</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        0.613605  0.437802         0.309835      0.745854   \n",
       "1  train        0.636850  0.455452         0.327352      0.748263   \n",
       "2    val        0.622809  0.436274         0.313700      0.716067   \n",
       "0   test        0.624088  0.434716         0.312984      0.711409   \n",
       "1   test        0.596180  0.442513         0.305231      0.804224   \n",
       "\n",
       "   average_precision_score  roc_auc_score  dset_naics_holdout  \n",
       "0                 0.346552       0.709198                 NaN  \n",
       "1                 0.381655       0.734007                 NaN  \n",
       "2                 0.351811       0.708278                 NaN  \n",
       "0                 0.346789       0.704898                 0.0  \n",
       "1                 0.348779       0.719503                 1.0  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics_dset_df = all_predictions_df.groupby('dset') \\\n",
    "    .apply(lambda x: sbnn_metrics.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: sbnn_metrics.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": "code",
   "execution_count": null,
   "id": "52e382b5-e9dc-47a1-9bf4-6d0b6b4a0c61",
   "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
}