{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Attempting a permutation approach based on matching demographics in users and comparing the mask based detertminants in the citizen shield project"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### Discussion\n",
    "- Main point is that the evolutionary potential of a system depends on its current state\n",
    "- I'm considering applications to psychology/behaviour\n",
    "- Mainly: a dataset has people as rows and features as columns, Likert scale. Let's say the columns are three behaviours and five demographics.\n",
    "- For every row, I want to find the most similar row in terms of demographics and the combination of behaviours.\n",
    "- Let's say more behaviours are better, so I want to find a slightly superior comparison person to everyone, especially for those low in the behaviours\n",
    "- So in theory, the person could be nudged to the superior 3-dimensional behavioural state, even though their 5-dimensional demographics are unchangable\n",
    "- (Three and five coming from here)\n",
    "- ... and least energy cost would be a switch to the most adjacent state\n",
    "- A new variable would be, for every person, the distance to their closest adjacent superior\n",
    "- Actually, you could just group the data by permutations of the 5 demographics, for each take a distance matrix of the 3 behaviours, and -- for each row -- pull the adjacent row numbers which are not identical\n",
    "- Is this a stupid idea?\n",
    "- Could then look at distance distributions of the groups (i.e. permutations of the 5 demographics), and a uniform distribution would be the nicest -- everyone in that group would have an improved version of themselves, so we \"know\" based on the data, that it's not impossible for someone with those demographics to improve behaviours\n",
    "- A bimodal distribution would be problematic, as there'd be a potentially uncrossable chasm for the people in the lower behavioural mode\n",
    "- So interventions could be focused on those groups, where individuals have potential for large stepwise improvement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from jmspack.utils import JmsColors, flatten\n",
    "from itertools import product\n",
    "from scipy.spatial.distance import pdist, squareform\n",
    "from pyxdameraulevenshtein import damerau_levenshtein_distance as DLD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "if \"jms_style_sheet\" in plt.style.available:\n",
    "    _ = plt.style.use(\"jms_style_sheet\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = (pd.read_csv(\"kp_determinants2/kp_determinants_and_demographics.csv\")\n",
    "      .dropna()\n",
    "     )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 2407 entries, 0 to 2458\n",
      "Data columns (total 31 columns):\n",
      " #   Column                              Non-Null Count  Dtype  \n",
      "---  ------                              --------------  -----  \n",
      " 0   demographic_gender                  2407 non-null   int64  \n",
      " 1   demographic_age                     2407 non-null   int64  \n",
      " 2   demographic_region                  2407 non-null   float64\n",
      " 3   demographic_education               2407 non-null   float64\n",
      " 4   demographic_living_with             2407 non-null   float64\n",
      " 5   demographic_underage_children       2407 non-null   object \n",
      " 6   demographic_income                  2407 non-null   float64\n",
      " 7   effectiveness_masks                 2407 non-null   int64  \n",
      " 8   effectiveness_hometests             2407 non-null   int64  \n",
      " 9   effectiveness_ventilation           2407 non-null   int64  \n",
      " 10  effectiveness_quarantine            2407 non-null   int64  \n",
      " 11  effectiveness_avoid_meeting         2407 non-null   int64  \n",
      " 12  effectiveness_announce_test_result  2407 non-null   int64  \n",
      " 13  effectiveness_avoid_public_events   2407 non-null   int64  \n",
      " 14  effectiveness_covid_passport        2407 non-null   int64  \n",
      " 15  ease_masks                          2407 non-null   int64  \n",
      " 16  ease_hometests                      2407 non-null   int64  \n",
      " 17  ease_ventilation                    2407 non-null   int64  \n",
      " 18  ease_quarantine                     2407 non-null   int64  \n",
      " 19  ease_avoid_meeting                  2407 non-null   int64  \n",
      " 20  ease_announce_test_result           2407 non-null   int64  \n",
      " 21  ease_avoid_public_events            2407 non-null   int64  \n",
      " 22  ease_covid_passport                 2407 non-null   int64  \n",
      " 23  persistence_masks                   2407 non-null   int64  \n",
      " 24  persistence_hometests               2407 non-null   int64  \n",
      " 25  persistence_ventilation             2407 non-null   int64  \n",
      " 26  persistence_quarantine              2407 non-null   int64  \n",
      " 27  persistence_avoid_meeting           2407 non-null   int64  \n",
      " 28  persistence_announce_test_result    2407 non-null   int64  \n",
      " 29  persistence_avoid_public_events     2407 non-null   int64  \n",
      " 30  persistence_covid_passport          2407 non-null   int64  \n",
      "dtypes: float64(4), int64(26), object(1)\n",
      "memory usage: 601.8+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "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>demographic_gender</th>\n",
       "      <th>demographic_age</th>\n",
       "      <th>demographic_region</th>\n",
       "      <th>demographic_education</th>\n",
       "      <th>demographic_living_with</th>\n",
       "      <th>demographic_underage_children</th>\n",
       "      <th>demographic_income</th>\n",
       "      <th>effectiveness_masks</th>\n",
       "      <th>effectiveness_hometests</th>\n",
       "      <th>effectiveness_ventilation</th>\n",
       "      <th>...</th>\n",
       "      <th>ease_avoid_public_events</th>\n",
       "      <th>ease_covid_passport</th>\n",
       "      <th>persistence_masks</th>\n",
       "      <th>persistence_hometests</th>\n",
       "      <th>persistence_ventilation</th>\n",
       "      <th>persistence_quarantine</th>\n",
       "      <th>persistence_avoid_meeting</th>\n",
       "      <th>persistence_announce_test_result</th>\n",
       "      <th>persistence_avoid_public_events</th>\n",
       "      <th>persistence_covid_passport</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>17.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>17.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>17.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>17.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>17.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>No</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   demographic_gender  demographic_age  demographic_region  \\\n",
       "0                   1                7                17.0   \n",
       "1                   1                4                17.0   \n",
       "2                   1                6                17.0   \n",
       "3                   1                7                17.0   \n",
       "4                   1                4                17.0   \n",
       "\n",
       "   demographic_education  demographic_living_with  \\\n",
       "0                    2.0                      2.0   \n",
       "1                    3.0                      2.0   \n",
       "2                    4.0                      2.0   \n",
       "3                    4.0                      2.0   \n",
       "4                    2.0                      1.0   \n",
       "\n",
       "  demographic_underage_children  demographic_income  effectiveness_masks  \\\n",
       "0                           Yes                 3.0                    2   \n",
       "1                           Yes                 2.0                    3   \n",
       "2                           Yes                 3.0                    5   \n",
       "3                           Yes                 3.0                    5   \n",
       "4                            No                 4.0                    1   \n",
       "\n",
       "   effectiveness_hometests  effectiveness_ventilation  ...  \\\n",
       "0                        1                          5  ...   \n",
       "1                        1                          5  ...   \n",
       "2                        5                          5  ...   \n",
       "3                        5                          4  ...   \n",
       "4                        1                          3  ...   \n",
       "\n",
       "   ease_avoid_public_events  ease_covid_passport  persistence_masks  \\\n",
       "0                         5                    0                  3   \n",
       "1                         5                    5                  2   \n",
       "2                         2                    4                  3   \n",
       "3                         2                    5                  5   \n",
       "4                         1                    1                  1   \n",
       "\n",
       "   persistence_hometests  persistence_ventilation  persistence_quarantine  \\\n",
       "0                      4                        5                       4   \n",
       "1                      1                        5                       1   \n",
       "2                      4                        5                       3   \n",
       "3                      5                        5                       5   \n",
       "4                      1                        1                       1   \n",
       "\n",
       "   persistence_avoid_meeting  persistence_announce_test_result  \\\n",
       "0                          4                                 5   \n",
       "1                          1                                 5   \n",
       "2                          3                                 5   \n",
       "3                          3                                 5   \n",
       "4                          1                                 1   \n",
       "\n",
       "   persistence_avoid_public_events  persistence_covid_passport  \n",
       "0                                5                           0  \n",
       "1                                4                           5  \n",
       "2                                1                           4  \n",
       "3                                3                           5  \n",
       "4                                1                           1  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2407, 31)\n"
     ]
    }
   ],
   "source": [
    "display(df.head()); print(df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "7f512da2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# demo_columns = df.filter(regex=\"demographic\").drop([\"demographic_region\", \"demographic_age\"], axis=1).columns.tolist()\n",
    "# demo_columns = df.filter(regex=\"demographic\").drop([\"demographic_region\"], axis=1).columns.tolist()\n",
    "demo_columns = [\"demographic_region\", \"demographic_age\"]\n",
    "# feature_columns = df.filter(regex=\"ease\").columns.tolist()\n",
    "feature_columns = df.filter(regex=\"effectiveness\").columns.tolist()[0:3]\n",
    "# feature_columns = [\"effectiveness_ventilation\"]\n",
    "# feature_columns = df.filter(regex=\"persistence\").columns.tolist()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "f28d9e11",
   "metadata": {},
   "source": [
    "### Create a demo_options_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "26924e3d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "demographic_region    18\n",
       "demographic_age       12\n",
       "dtype: int64"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[demo_columns].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "9f1bb03f",
   "metadata": {},
   "outputs": [],
   "source": [
    "options_list = [list(x) for x in df[demo_columns].apply(lambda col: col.unique()).tolist()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "30c70db6",
   "metadata": {},
   "outputs": [],
   "source": [
    "all_options_list = list(product(*options_list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "6e3d91bd",
   "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>option_number</th>\n",
       "      <th>demographic_region</th>\n",
       "      <th>demographic_age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>17.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>17.0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>17.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>17.0</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   option_number  demographic_region  demographic_age\n",
       "0              0                17.0                7\n",
       "1              1                17.0                4\n",
       "2              2                17.0                6\n",
       "3              3                17.0                3\n",
       "4              4                17.0               11"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(216, 3)\n"
     ]
    }
   ],
   "source": [
    "options_df = (pd.DataFrame(all_options_list, columns=demo_columns)\n",
    "              .reset_index()\n",
    "              .rename(columns={\"index\": \"option_number\"}))\n",
    "display(options_df.head()); print(options_df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "5de665d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_perm = pd.DataFrame()\n",
    "for row in options_df.index:\n",
    "    tmp = pd.merge(options_df.iloc[[row], :], df[demo_columns + feature_columns], on=demo_columns)\n",
    "    df_perm = pd.concat([df_perm, tmp])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "c34090ba",
   "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>option_number</th>\n",
       "      <th>frequency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>91</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   option_number  frequency\n",
       "0              8         97\n",
       "1              4         91"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(26, 2)\n"
     ]
    }
   ],
   "source": [
    "amount_cutoff = 20\n",
    "amount_top_demos = 10\n",
    "plot_df = (df_perm[\"option_number\"]\n",
    ".value_counts()\n",
    ".to_frame()\n",
    ".reset_index()\n",
    ".rename(columns={\"index\": \"option_number\", \"option_number\": \"frequency\"})\n",
    ".pipe(lambda d: d[d[\"frequency\"]>=amount_cutoff])\n",
    ")\n",
    "top_demos_list = plot_df[\"option_number\"].head(amount_top_demos).values.tolist()\n",
    "display(plot_df.head(2)); print(plot_df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "f8c1c6e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = plt.figure(figsize=(2, 8))\n",
    "_ = sns.heatmap(data=plot_df.set_index(\"option_number\"), cmap=\"Blues\", annot=True, fmt=\".0f\", cbar=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "9196dbba",
   "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>option_number</th>\n",
       "      <th>demographic_region</th>\n",
       "      <th>demographic_age</th>\n",
       "      <th>effectiveness_masks</th>\n",
       "      <th>effectiveness_hometests</th>\n",
       "      <th>effectiveness_ventilation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>17.0</td>\n",
       "      <td>11</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>17.0</td>\n",
       "      <td>11</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  option_number  demographic_region  demographic_age  effectiveness_masks  \\\n",
       "0             4                17.0               11                    5   \n",
       "1             4                17.0               11                    4   \n",
       "\n",
       "   effectiveness_hometests  effectiveness_ventilation  \n",
       "0                        3                          0  \n",
       "1                        3                          3  "
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_df = (df_perm.pipe(lambda d: d[d[\"option_number\"].isin([top_demos_list[0], top_demos_list[1]])])\n",
    "           .assign(**{\"option_number\": lambda d: d[\"option_number\"].astype(\"str\")}))\n",
    "plot_df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "9a1d9a3a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = plt.figure(figsize=(30, 5))\n",
    "_ = sns.swarmplot(data=plot_df, x=feature_columns[2], y=feature_columns[1], hue=\"option_number\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "id": "6d34aa34",
   "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>feature_diff_mean</th>\n",
       "      <th>option_number</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.666667</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>-1.666667</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>0.333333</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    feature_diff_mean option_number\n",
       "61           1.000000             0\n",
       "0            0.666667             0\n",
       "27          -1.666667             0\n",
       "9            1.000000             0\n",
       "24           0.333333             0"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selection_df = df_perm.pipe(lambda d: d[d[\"option_number\"].isin(top_demos_list)])\n",
    "plot_df = (selection_df\n",
    " .drop(demo_columns, axis=1)\n",
    " .sort_values(by=[\"option_number\"] + feature_columns)\n",
    " .groupby(\"option_number\")\n",
    " .diff(axis=0)\n",
    " .mean(axis=1)\n",
    " .to_frame(\"feature_diff_mean\")\n",
    " .assign(**{\"option_number\": selection_df[\"option_number\"].values})\n",
    "  .dropna()\n",
    " .assign(**{\"option_number\": lambda d: d[\"option_number\"].astype(\"category\")})\n",
    " )\n",
    "plot_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "id": "51a2a776",
   "metadata": {},
   "outputs": [],
   "source": [
    "if amount_top_demos <= 10:\n",
    "    show_legend=True\n",
    "else:\n",
    "    show_legend=False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "efffbd37",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = plt.figure(figsize=(15, 5))\n",
    "_ = sns.histplot(data=plot_df, \n",
    "                 x=\"feature_diff_mean\", \n",
    "                 hue=\"option_number\", \n",
    "                 kde=True, bins=50, \n",
    "                 legend=show_legend)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "id": "d6e09516",
   "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>option_number</th>\n",
       "      <th>variable</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>effectiveness_masks</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>effectiveness_masks</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>effectiveness_masks</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>effectiveness_masks</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>effectiveness_masks</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  option_number             variable  value\n",
       "0             0  effectiveness_masks    0.0\n",
       "1             0  effectiveness_masks    1.0\n",
       "2             0  effectiveness_masks    1.0\n",
       "3             0  effectiveness_masks    0.0\n",
       "4             0  effectiveness_masks    0.0"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_df = (selection_df\n",
    " .drop(demo_columns, axis=1)\n",
    " .sort_values(by=[\"option_number\"] + feature_columns)\n",
    " .groupby(\"option_number\")\n",
    " .diff(axis=0)\n",
    " .assign(**{\"option_number\": selection_df[\"option_number\"].values})\n",
    "  .dropna()\n",
    " .assign(**{\"option_number\": lambda d: d[\"option_number\"].astype(\"category\")})\n",
    " .melt(id_vars=\"option_number\")\n",
    " )\n",
    "plot_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "id": "4e6e4061",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = plt.figure(figsize=(15, 5))\n",
    "_ = sns.stripplot(data=plot_df, \n",
    "                  x=\"variable\", \n",
    "                  y=\"value\", \n",
    "                  hue=\"option_number\", \n",
    "                  dodge=True, \n",
    "                  legend=show_legend)\n",
    "_ = plt.xticks(rotation=90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "id": "6a5d23d3",
   "metadata": {},
   "outputs": [],
   "source": [
    "DLD_df = pd.DataFrame()\n",
    "for option in top_demos_list:\n",
    "    tmp_df = (selection_df\n",
    "    .pipe(lambda d: d[d[\"option_number\"]==option])\n",
    "    .assign(**{\"combined_score\": lambda d: d[feature_columns]\n",
    "                .astype(str)\n",
    "                .sum(axis=1)\n",
    "                .astype(int)})\n",
    "    .loc[:, [\"option_number\", \"combined_score\"]]\n",
    "    .sort_values(by=[\"option_number\", \"combined_score\"])\n",
    "    .assign(**{\"combined_score\": lambda d: d[\"combined_score\"].astype(str)})\n",
    "    .reset_index(drop=True)\n",
    "    )\n",
    "    tmp_df[\"DLD\"] = [np.nan] + [DLD(tmp_df.loc[x, \"combined_score\"], \n",
    "                    tmp_df.loc[x+1, \"combined_score\"]) for x in tmp_df.index.tolist()[:-1]]\n",
    "    DLD_df = pd.concat([DLD_df, tmp_df])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "id": "4e613b59",
   "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>option_number</th>\n",
       "      <th>combined_score</th>\n",
       "      <th>DLD</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8</td>\n",
       "      <td>143</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8</td>\n",
       "      <td>150</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8</td>\n",
       "      <td>215</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8</td>\n",
       "      <td>225</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>3</td>\n",
       "      <td>540</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>3</td>\n",
       "      <td>543</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>3</td>\n",
       "      <td>545</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>3</td>\n",
       "      <td>555</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>3</td>\n",
       "      <td>555</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>754 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    option_number combined_score  DLD\n",
       "0               8              0  NaN\n",
       "1               8            143  3.0\n",
       "2               8            150  2.0\n",
       "3               8            215  2.0\n",
       "4               8            225  1.0\n",
       "..            ...            ...  ...\n",
       "42              3            540  2.0\n",
       "43              3            543  1.0\n",
       "44              3            545  1.0\n",
       "45              3            555  1.0\n",
       "46              3            555  0.0\n",
       "\n",
       "[754 rows x 3 columns]"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DLD_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "id": "93c8af8a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = plt.figure(figsize=(5, 5))\n",
    "_ = sns.histplot(data=DLD_df.dropna().assign(**{\"option_number\": lambda d: d[\"option_number\"].astype(str)}), \n",
    "                 x=\"DLD\", \n",
    "                 hue=\"option_number\", \n",
    "                 kde=True, bins=50, \n",
    "                 legend=show_legend)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6387f455",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ds_env",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.4"
  },
  "vscode": {
   "interpreter": {
    "hash": "179735f5087b26bd07935103d9bc58d14ac417e85d5e87ba282c2f2ba7fe1bd1"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}