{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "1ccb3bcc",
   "metadata": {},
   "source": [
    "# Matching Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31614f02",
   "metadata": {},
   "source": [
    "In this notebook, we demonstrate the MatchingData class, which organizes population data for matching, and some plotting tools for visualizing the data. You can download this notebook to run yourself here: https://github.com/Bayer-Group/pybalance/blob/main/sphinx/demos/matching_data.ipynb."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4e60274e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os \n",
    "import logging \n",
    "logging.basicConfig(\n",
    "    format=\"%(levelname)-4s [%(filename)s:%(lineno)d] %(message)s\",\n",
    "    level='INFO',\n",
    ")\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "import pybalance\n",
    "from pybalance import MatchingData, MatchingHeaders, split_target_pool\n",
    "from pybalance.visualization import (\n",
    "    plot_numeric_features, \n",
    "    plot_categoric_features, \n",
    "    plot_binary_features,\n",
    "    plot_joint_numeric_distributions,\n",
    "    plot_joint_numeric_categoric_distributions,\n",
    "    plot_per_feature_loss\n",
    ")\n",
    "from pybalance.sim import get_paper_dataset_path"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dda436e5",
   "metadata": {},
   "source": [
    "## Initializing MatchingData"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5184f024",
   "metadata": {},
   "source": [
    "The MatchingData class is a thin wrapper around pandas DataFrame that additionally keeps track of certain metadata about the columns relevant for matching. MatchingData can be initialized from either a string or pandas DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "02250cfb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <b>Headers Numeric: </b><br>\n",
       "        ['age', 'height', 'weight']<br><br>\n",
       "        <b>Headers Categoric: </b><br>\n",
       "        ['gender', 'haircolor', 'country', 'binary_0', 'binary_1', 'binary_2', 'binary_3'] <br><br>\n",
       "        <b>Populations</b> <br>\n",
       "        ['pool', 'target'] <br>\n",
       "        <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>age</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>gender</th>\n",
       "      <th>haircolor</th>\n",
       "      <th>country</th>\n",
       "      <th>population</th>\n",
       "      <th>binary_0</th>\n",
       "      <th>binary_1</th>\n",
       "      <th>binary_2</th>\n",
       "      <th>binary_3</th>\n",
       "      <th>patient_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>64.854093</td>\n",
       "      <td>189.466850</td>\n",
       "      <td>88.835049</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>135740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>52.571993</td>\n",
       "      <td>158.134940</td>\n",
       "      <td>94.215107</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>49288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25.828361</td>\n",
       "      <td>154.692482</td>\n",
       "      <td>94.226222</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>256676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>70.177571</td>\n",
       "      <td>160.536632</td>\n",
       "      <td>94.244356</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>338287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>73.779164</td>\n",
       "      <td>153.551419</td>\n",
       "      <td>86.161814</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>72849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274995</th>\n",
       "      <td>62.547794</td>\n",
       "      <td>186.005015</td>\n",
       "      <td>50.975051</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>579081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274996</th>\n",
       "      <td>69.879934</td>\n",
       "      <td>142.371386</td>\n",
       "      <td>100.138389</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>569939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274997</th>\n",
       "      <td>56.921402</td>\n",
       "      <td>130.639589</td>\n",
       "      <td>108.745182</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>532419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274998</th>\n",
       "      <td>34.082754</td>\n",
       "      <td>174.764051</td>\n",
       "      <td>67.998396</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>566266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274999</th>\n",
       "      <td>60.981259</td>\n",
       "      <td>137.419436</td>\n",
       "      <td>89.897817</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>target</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>544231</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>275000 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "<pybalance.utils.matching_data.MatchingData at 0x7f7b5821e850>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# initialize MatchingData from path\n",
    "data_path = get_paper_dataset_path()\n",
    "m = MatchingData(data=data_path)\n",
    "m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a45adfde",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <b>Headers Numeric: </b><br>\n",
       "        ['age', 'height', 'weight']<br><br>\n",
       "        <b>Headers Categoric: </b><br>\n",
       "        ['gender', 'haircolor', 'country', 'binary_0', 'binary_1', 'binary_2', 'binary_3'] <br><br>\n",
       "        <b>Populations</b> <br>\n",
       "        ['pool', 'target'] <br>\n",
       "        <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>age</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>gender</th>\n",
       "      <th>haircolor</th>\n",
       "      <th>country</th>\n",
       "      <th>population</th>\n",
       "      <th>binary_0</th>\n",
       "      <th>binary_1</th>\n",
       "      <th>binary_2</th>\n",
       "      <th>binary_3</th>\n",
       "      <th>patient_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>64.854093</td>\n",
       "      <td>189.466850</td>\n",
       "      <td>88.835049</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>135740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>52.571993</td>\n",
       "      <td>158.134940</td>\n",
       "      <td>94.215107</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>49288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25.828361</td>\n",
       "      <td>154.692482</td>\n",
       "      <td>94.226222</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>256676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>70.177571</td>\n",
       "      <td>160.536632</td>\n",
       "      <td>94.244356</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>338287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>73.779164</td>\n",
       "      <td>153.551419</td>\n",
       "      <td>86.161814</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>72849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274995</th>\n",
       "      <td>62.547794</td>\n",
       "      <td>186.005015</td>\n",
       "      <td>50.975051</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>579081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274996</th>\n",
       "      <td>69.879934</td>\n",
       "      <td>142.371386</td>\n",
       "      <td>100.138389</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>569939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274997</th>\n",
       "      <td>56.921402</td>\n",
       "      <td>130.639589</td>\n",
       "      <td>108.745182</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>532419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274998</th>\n",
       "      <td>34.082754</td>\n",
       "      <td>174.764051</td>\n",
       "      <td>67.998396</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>566266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274999</th>\n",
       "      <td>60.981259</td>\n",
       "      <td>137.419436</td>\n",
       "      <td>89.897817</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>target</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>544231</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>275000 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "<pybalance.utils.matching_data.MatchingData at 0x7f7b81fa35e0>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# initialize MatchingData from pandas DataFrame\n",
    "data = pd.read_parquet(data_path)\n",
    "m = MatchingData(data=data)\n",
    "m"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51980126",
   "metadata": {},
   "source": [
    "MatchingData will infer which covariates to use for matching and the separation of these into numeric and categoric, unless explicitly specified. Here we specify a subset of the covariates to use for matching. Note that the unused columns are still present in the data, but will simply not be used for matching."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4099f276",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <b>Headers Numeric: </b><br>\n",
       "        ['age', 'weight', 'height']<br><br>\n",
       "        <b>Headers Categoric: </b><br>\n",
       "        ['country', 'gender', 'binary_0', 'binary_1'] <br><br>\n",
       "        <b>Populations</b> <br>\n",
       "        ['pool', 'target'] <br>\n",
       "        <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>age</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>gender</th>\n",
       "      <th>haircolor</th>\n",
       "      <th>country</th>\n",
       "      <th>population</th>\n",
       "      <th>binary_0</th>\n",
       "      <th>binary_1</th>\n",
       "      <th>binary_2</th>\n",
       "      <th>binary_3</th>\n",
       "      <th>patient_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>64.854093</td>\n",
       "      <td>189.466850</td>\n",
       "      <td>88.835049</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>135740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>52.571993</td>\n",
       "      <td>158.134940</td>\n",
       "      <td>94.215107</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>49288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25.828361</td>\n",
       "      <td>154.692482</td>\n",
       "      <td>94.226222</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>256676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>70.177571</td>\n",
       "      <td>160.536632</td>\n",
       "      <td>94.244356</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>338287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>73.779164</td>\n",
       "      <td>153.551419</td>\n",
       "      <td>86.161814</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>72849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274995</th>\n",
       "      <td>62.547794</td>\n",
       "      <td>186.005015</td>\n",
       "      <td>50.975051</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>579081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274996</th>\n",
       "      <td>69.879934</td>\n",
       "      <td>142.371386</td>\n",
       "      <td>100.138389</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>569939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274997</th>\n",
       "      <td>56.921402</td>\n",
       "      <td>130.639589</td>\n",
       "      <td>108.745182</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>532419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274998</th>\n",
       "      <td>34.082754</td>\n",
       "      <td>174.764051</td>\n",
       "      <td>67.998396</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>566266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274999</th>\n",
       "      <td>60.981259</td>\n",
       "      <td>137.419436</td>\n",
       "      <td>89.897817</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>target</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>544231</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>275000 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "<pybalance.utils.matching_data.MatchingData at 0x7f7b48e59eb0>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "headers = MatchingHeaders(\n",
    "    categoric=['country', 'gender', 'binary_0', 'binary_1'],\n",
    "    numeric=['age', 'weight', 'height']\n",
    ")\n",
    "m_restricted_features = MatchingData(\n",
    "    data=data, \n",
    "    headers=headers\n",
    ")\n",
    "m_restricted_features"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d101d154",
   "metadata": {},
   "source": [
    "## Exploring MatchingData"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39ecd310",
   "metadata": {},
   "source": [
    "The describe*() methods can be used to generate summary tables of the matching covariates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2122eccd",
   "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></th>\n",
       "      <th>pool</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>population size</th>\n",
       "      <th>N</th>\n",
       "      <td>250000.00</td>\n",
       "      <td>25000.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">gender</th>\n",
       "      <th>0.0</th>\n",
       "      <td>120054.00</td>\n",
       "      <td>12956.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>129946.00</td>\n",
       "      <td>12044.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">haircolor</th>\n",
       "      <th>0.0</th>\n",
       "      <td>100096.00</td>\n",
       "      <td>4924.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>75185.00</td>\n",
       "      <td>10055.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>74719.00</td>\n",
       "      <td>10021.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"6\" valign=\"top\">country</th>\n",
       "      <th>0.0</th>\n",
       "      <td>0.00</td>\n",
       "      <td>2490.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>25033.00</td>\n",
       "      <td>5045.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>49534.00</td>\n",
       "      <td>4981.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>75337.00</td>\n",
       "      <td>2474.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>74934.00</td>\n",
       "      <td>5010.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>25162.00</td>\n",
       "      <td>5000.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">binary_0</th>\n",
       "      <th>0.0</th>\n",
       "      <td>225028.00</td>\n",
       "      <td>17535.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>24972.00</td>\n",
       "      <td>7465.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">binary_1</th>\n",
       "      <th>0.0</th>\n",
       "      <td>175673.00</td>\n",
       "      <td>12527.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>74327.00</td>\n",
       "      <td>12473.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">binary_2</th>\n",
       "      <th>0.0</th>\n",
       "      <td>125113.00</td>\n",
       "      <td>17472.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>124887.00</td>\n",
       "      <td>7528.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">binary_3</th>\n",
       "      <th>0.0</th>\n",
       "      <td>49933.00</td>\n",
       "      <td>12562.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>200067.00</td>\n",
       "      <td>12438.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"7\" valign=\"top\">age</th>\n",
       "      <th>mean</th>\n",
       "      <td>55.27</td>\n",
       "      <td>48.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>13.18</td>\n",
       "      <td>14.39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>18.01</td>\n",
       "      <td>18.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>q25</th>\n",
       "      <td>46.38</td>\n",
       "      <td>37.29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>median</th>\n",
       "      <td>57.15</td>\n",
       "      <td>48.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>q75</th>\n",
       "      <td>66.10</td>\n",
       "      <td>59.85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>75.00</td>\n",
       "      <td>75.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"7\" valign=\"top\">height</th>\n",
       "      <th>mean</th>\n",
       "      <td>159.13</td>\n",
       "      <td>153.68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>19.84</td>\n",
       "      <td>16.45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>125.00</td>\n",
       "      <td>125.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>q25</th>\n",
       "      <td>142.09</td>\n",
       "      <td>140.29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>median</th>\n",
       "      <td>158.74</td>\n",
       "      <td>152.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>q75</th>\n",
       "      <td>175.87</td>\n",
       "      <td>165.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>195.00</td>\n",
       "      <td>195.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"7\" valign=\"top\">weight</th>\n",
       "      <th>mean</th>\n",
       "      <td>88.30</td>\n",
       "      <td>82.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>16.32</td>\n",
       "      <td>18.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>50.00</td>\n",
       "      <td>50.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>q25</th>\n",
       "      <td>76.39</td>\n",
       "      <td>66.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>median</th>\n",
       "      <td>88.85</td>\n",
       "      <td>81.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>q75</th>\n",
       "      <td>100.88</td>\n",
       "      <td>97.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>120.00</td>\n",
       "      <td>120.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             pool    target\n",
       "population size N       250000.00  25000.00\n",
       "gender          0.0     120054.00  12956.00\n",
       "                1.0     129946.00  12044.00\n",
       "haircolor       0.0     100096.00   4924.00\n",
       "                1.0      75185.00  10055.00\n",
       "                2        74719.00  10021.00\n",
       "country         0.0          0.00   2490.00\n",
       "                1.0      25033.00   5045.00\n",
       "                2        49534.00   4981.00\n",
       "                3        75337.00   2474.00\n",
       "                4        74934.00   5010.00\n",
       "                5        25162.00   5000.00\n",
       "binary_0        0.0     225028.00  17535.00\n",
       "                1.0      24972.00   7465.00\n",
       "binary_1        0.0     175673.00  12527.00\n",
       "                1.0      74327.00  12473.00\n",
       "binary_2        0.0     125113.00  17472.00\n",
       "                1.0     124887.00   7528.00\n",
       "binary_3        0.0      49933.00  12562.00\n",
       "                1.0     200067.00  12438.00\n",
       "age             mean        55.27     48.33\n",
       "                std         13.18     14.39\n",
       "                min         18.01     18.01\n",
       "                q25         46.38     37.29\n",
       "                median      57.15     48.74\n",
       "                q75         66.10     59.85\n",
       "                max         75.00     75.00\n",
       "height          mean       159.13    153.68\n",
       "                std         19.84     16.45\n",
       "                min        125.00    125.00\n",
       "                q25        142.09    140.29\n",
       "                median     158.74    152.75\n",
       "                q75        175.87    165.95\n",
       "                max        195.00    195.00\n",
       "weight          mean        88.30     82.25\n",
       "                std         16.32     18.89\n",
       "                min         50.00     50.00\n",
       "                q25         76.39     66.14\n",
       "                median      88.85     81.69\n",
       "                q75        100.88     97.41\n",
       "                max        120.00    120.00"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m.describe(normalize=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "50da290d",
   "metadata": {},
   "source": [
    "You can access fields on the underlying data similarly to how you would in pandas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5401a9cc",
   "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>population</th>\n",
       "      <th>gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>pool</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>pool</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>pool</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>pool</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>pool</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274995</th>\n",
       "      <td>target</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274996</th>\n",
       "      <td>target</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274997</th>\n",
       "      <td>target</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274998</th>\n",
       "      <td>target</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274999</th>\n",
       "      <td>target</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>275000 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       population  gender\n",
       "0            pool     1.0\n",
       "1            pool     1.0\n",
       "2            pool     1.0\n",
       "3            pool     1.0\n",
       "4            pool     0.0\n",
       "...           ...     ...\n",
       "274995     target     0.0\n",
       "274996     target     1.0\n",
       "274997     target     1.0\n",
       "274998     target     0.0\n",
       "274999     target     1.0\n",
       "\n",
       "[275000 rows x 2 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m[['population', 'gender']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "191d7c65",
   "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>age</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>gender</th>\n",
       "      <th>haircolor</th>\n",
       "      <th>country</th>\n",
       "      <th>population</th>\n",
       "      <th>binary_0</th>\n",
       "      <th>binary_1</th>\n",
       "      <th>binary_2</th>\n",
       "      <th>binary_3</th>\n",
       "      <th>patient_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>73.779164</td>\n",
       "      <td>153.551419</td>\n",
       "      <td>86.161814</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>72849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>67.404918</td>\n",
       "      <td>132.383184</td>\n",
       "      <td>67.107753</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>171211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>61.489148</td>\n",
       "      <td>140.780034</td>\n",
       "      <td>73.662572</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>20695</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>73.718093</td>\n",
       "      <td>133.743721</td>\n",
       "      <td>58.879321</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>58718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>70.707782</td>\n",
       "      <td>156.629048</td>\n",
       "      <td>70.681391</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>352801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274990</th>\n",
       "      <td>19.063519</td>\n",
       "      <td>167.704149</td>\n",
       "      <td>59.876565</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>536365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274992</th>\n",
       "      <td>58.745450</td>\n",
       "      <td>146.747313</td>\n",
       "      <td>70.291448</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>535279</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274993</th>\n",
       "      <td>55.736083</td>\n",
       "      <td>132.434020</td>\n",
       "      <td>92.264209</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>595582</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274995</th>\n",
       "      <td>62.547794</td>\n",
       "      <td>186.005015</td>\n",
       "      <td>50.975051</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>579081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274998</th>\n",
       "      <td>34.082754</td>\n",
       "      <td>174.764051</td>\n",
       "      <td>67.998396</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>566266</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>133010 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              age      height     weight  gender  haircolor  country  \\\n",
       "4       73.779164  153.551419  86.161814     0.0          0        1   \n",
       "7       67.404918  132.383184  67.107753     0.0          0        5   \n",
       "11      61.489148  140.780034  73.662572     0.0          0        1   \n",
       "12      73.718093  133.743721  58.879321     0.0          1        4   \n",
       "13      70.707782  156.629048  70.681391     0.0          0        2   \n",
       "...           ...         ...        ...     ...        ...      ...   \n",
       "274990  19.063519  167.704149  59.876565     0.0          2        4   \n",
       "274992  58.745450  146.747313  70.291448     0.0          2        4   \n",
       "274993  55.736083  132.434020  92.264209     0.0          1        4   \n",
       "274995  62.547794  186.005015  50.975051     0.0          0        1   \n",
       "274998  34.082754  174.764051  67.998396     0.0          2        2   \n",
       "\n",
       "       population  binary_0  binary_1  binary_2  binary_3  patient_id  \n",
       "4            pool         0         0         1         1       72849  \n",
       "7            pool         0         0         0         1      171211  \n",
       "11           pool         0         0         0         1       20695  \n",
       "12           pool         0         0         1         0       58718  \n",
       "13           pool         0         1         1         1      352801  \n",
       "...           ...       ...       ...       ...       ...         ...  \n",
       "274990     target         0         1         1         1      536365  \n",
       "274992     target         0         0         0         1      535279  \n",
       "274993     target         0         1         0         0      595582  \n",
       "274995     target         0         0         1         1      579081  \n",
       "274998     target         0         0         0         1      566266  \n",
       "\n",
       "[133010 rows x 12 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m[m['gender'] == 0]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b63f8c72",
   "metadata": {},
   "source": [
    "Often our matching data consists of exactly two populations, a reference population, \n",
    "which we call the \"target\" and a population to be matched, which we call the \"pool\". \n",
    "It is sometimes convenient to split these two populations and the function \n",
    "split_target_pool does just that. The function will assign the smaller population to the\n",
    "target, unless explicitly given the name of the target population. Note that the returned\n",
    "values are pandas DataFrame objects and not MatchingData objects."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "e630b0f4",
   "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>age</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>gender</th>\n",
       "      <th>haircolor</th>\n",
       "      <th>country</th>\n",
       "      <th>population</th>\n",
       "      <th>binary_0</th>\n",
       "      <th>binary_1</th>\n",
       "      <th>binary_2</th>\n",
       "      <th>binary_3</th>\n",
       "      <th>patient_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>250000</th>\n",
       "      <td>57.266010</td>\n",
       "      <td>159.759575</td>\n",
       "      <td>94.325267</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>512966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250001</th>\n",
       "      <td>53.152645</td>\n",
       "      <td>145.515410</td>\n",
       "      <td>95.988094</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>540606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250002</th>\n",
       "      <td>34.079212</td>\n",
       "      <td>166.272208</td>\n",
       "      <td>73.090671</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>578266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250003</th>\n",
       "      <td>45.494927</td>\n",
       "      <td>144.336677</td>\n",
       "      <td>96.678251</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>target</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>559858</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250004</th>\n",
       "      <td>18.036012</td>\n",
       "      <td>174.843524</td>\n",
       "      <td>60.586475</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>588368</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              age      height     weight  gender  haircolor  country  \\\n",
       "250000  57.266010  159.759575  94.325267     0.0          1        4   \n",
       "250001  53.152645  145.515410  95.988094     0.0          1        2   \n",
       "250002  34.079212  166.272208  73.090671     0.0          1        2   \n",
       "250003  45.494927  144.336677  96.678251     1.0          2        5   \n",
       "250004  18.036012  174.843524  60.586475     0.0          1        2   \n",
       "\n",
       "       population  binary_0  binary_1  binary_2  binary_3  patient_id  \n",
       "250000     target         0         1         0         1      512966  \n",
       "250001     target         1         0         1         1      540606  \n",
       "250002     target         0         0         1         1      578266  \n",
       "250003     target         1         1         1         1      559858  \n",
       "250004     target         1         1         0         1      588368  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target, pool = split_target_pool(m)\n",
    "target.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "83f90e8a",
   "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>age</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>gender</th>\n",
       "      <th>haircolor</th>\n",
       "      <th>country</th>\n",
       "      <th>population</th>\n",
       "      <th>binary_0</th>\n",
       "      <th>binary_1</th>\n",
       "      <th>binary_2</th>\n",
       "      <th>binary_3</th>\n",
       "      <th>patient_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>250000</th>\n",
       "      <td>57.266010</td>\n",
       "      <td>159.759575</td>\n",
       "      <td>94.325267</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>512966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250001</th>\n",
       "      <td>53.152645</td>\n",
       "      <td>145.515410</td>\n",
       "      <td>95.988094</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>540606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250002</th>\n",
       "      <td>34.079212</td>\n",
       "      <td>166.272208</td>\n",
       "      <td>73.090671</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>578266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250003</th>\n",
       "      <td>45.494927</td>\n",
       "      <td>144.336677</td>\n",
       "      <td>96.678251</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>target</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>559858</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250004</th>\n",
       "      <td>18.036012</td>\n",
       "      <td>174.843524</td>\n",
       "      <td>60.586475</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>588368</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              age      height     weight  gender  haircolor  country  \\\n",
       "250000  57.266010  159.759575  94.325267     0.0          1        4   \n",
       "250001  53.152645  145.515410  95.988094     0.0          1        2   \n",
       "250002  34.079212  166.272208  73.090671     0.0          1        2   \n",
       "250003  45.494927  144.336677  96.678251     1.0          2        5   \n",
       "250004  18.036012  174.843524  60.586475     0.0          1        2   \n",
       "\n",
       "       population  binary_0  binary_1  binary_2  binary_3  patient_id  \n",
       "250000     target         0         1         0         1      512966  \n",
       "250001     target         1         0         1         1      540606  \n",
       "250002     target         0         0         1         1      578266  \n",
       "250003     target         1         1         1         1      559858  \n",
       "250004     target         1         1         0         1      588368  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target, pool = split_target_pool(m, target_name='pool')\n",
    "pool.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13d05376",
   "metadata": {},
   "source": [
    "## Visualizing MatchingData"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8dda104d",
   "metadata": {},
   "source": [
    "Some built-in tools help you get a quick visual snapshot of the data. Many of these plotting routines are thin wrappers around seaborn plotting routines with extra logic relevant to matching situations (e.g. where one of the populations is a reference population or where variables should be treated as numeric / categoric). In most cases, the user can pass along any keyword arguments that are understood by the underlying seaborn routine."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "029b22de",
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "888c6bc9-73c1-499b-bc0e-82801e46dbbb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <b>Headers Numeric: </b><br>\n",
       "        ['age', 'height', 'weight']<br><br>\n",
       "        <b>Headers Categoric: </b><br>\n",
       "        ['gender', 'haircolor', 'country', 'binary_0', 'binary_1', 'binary_2', 'binary_3'] <br><br>\n",
       "        <b>Populations</b> <br>\n",
       "        ['pool', 'target'] <br>\n",
       "        <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>age</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>gender</th>\n",
       "      <th>haircolor</th>\n",
       "      <th>country</th>\n",
       "      <th>population</th>\n",
       "      <th>binary_0</th>\n",
       "      <th>binary_1</th>\n",
       "      <th>binary_2</th>\n",
       "      <th>binary_3</th>\n",
       "      <th>patient_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>64.854093</td>\n",
       "      <td>189.466850</td>\n",
       "      <td>88.835049</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>135740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>52.571993</td>\n",
       "      <td>158.134940</td>\n",
       "      <td>94.215107</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>49288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25.828361</td>\n",
       "      <td>154.692482</td>\n",
       "      <td>94.226222</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>256676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>70.177571</td>\n",
       "      <td>160.536632</td>\n",
       "      <td>94.244356</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>338287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>73.779164</td>\n",
       "      <td>153.551419</td>\n",
       "      <td>86.161814</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>72849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274995</th>\n",
       "      <td>62.547794</td>\n",
       "      <td>186.005015</td>\n",
       "      <td>50.975051</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>579081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274996</th>\n",
       "      <td>69.879934</td>\n",
       "      <td>142.371386</td>\n",
       "      <td>100.138389</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>569939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274997</th>\n",
       "      <td>56.921402</td>\n",
       "      <td>130.639589</td>\n",
       "      <td>108.745182</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>532419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274998</th>\n",
       "      <td>34.082754</td>\n",
       "      <td>174.764051</td>\n",
       "      <td>67.998396</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>target</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>566266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274999</th>\n",
       "      <td>60.981259</td>\n",
       "      <td>137.419436</td>\n",
       "      <td>89.897817</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>target</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>544231</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>275000 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "<pybalance.utils.matching_data.MatchingData at 0x7f7b81fa35e0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "925b6b8c-abac-4956-969a-a6a7d4d33f9f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the standardized mean difference for each feature\n",
    "from pybalance.utils import BetaBalance\n",
    "bc = BetaBalance(m, standardize_difference=True)\n",
    "fig = plot_per_feature_loss(m, bc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "042b9a5a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x450 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plot_categoric_features(m, cumulative=False, palette='colorblind', include_binary=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "62efef9a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x900 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plot_numeric_features(m, bins=10, cumulative=False, palette='colorblind')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "19711749",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/sprivite/src/pybalance/venv/pybalance/lib/python3.9/site-packages/pybalance/visualization/distributions.py:331: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.\n",
      "  plt.gca().set_yticklabels([\"\"] * len(labels), minor=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x250 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plot_binary_features(m, palette='colorblind', orient_horizontal=False, standardize_difference=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pybalance",
   "language": "python",
   "name": "pybalance"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}