{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.05656,
     "end_time": "2020-05-08T03:34:32.329300",
     "exception": false,
     "start_time": "2020-05-08T03:34:32.272740",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# User Profile Data Prep\n",
    "\n",
    "This notebook does the data preparation for Bayesian inference for the model analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.05407,
     "end_time": "2020-05-08T03:34:32.438247",
     "exception": false,
     "start_time": "2020-05-08T03:34:32.384177",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "papermill": {
     "duration": 0.06991,
     "end_time": "2020-05-08T03:34:32.565188",
     "exception": false,
     "start_time": "2020-05-08T03:34:32.495278",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from pathlib import Path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "papermill": {
     "duration": 2.090999,
     "end_time": "2020-05-08T03:34:34.711129",
     "exception": false,
     "start_time": "2020-05-08T03:34:32.620130",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from plotnine import *\n",
    "import ujson"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "papermill": {
     "duration": 0.194735,
     "end_time": "2020-05-08T03:34:34.960602",
     "exception": false,
     "start_time": "2020-05-08T03:34:34.765867",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from bookgender.config import data_dir\n",
    "from bookgender.nbutils import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "papermill": {
     "duration": 0.076641,
     "end_time": "2020-05-08T03:34:35.092662",
     "exception": false,
     "start_time": "2020-05-08T03:34:35.016021",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "using figure dir figures/ProfileData\n"
     ]
    }
   ],
   "source": [
    "fig_dir = init_figs('ProfileData')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.055835,
     "end_time": "2020-05-08T03:34:35.204268",
     "exception": false,
     "start_time": "2020-05-08T03:34:35.148433",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Load Data\n",
    "\n",
    "We need to load the author-gender information:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "papermill": {
     "duration": 7.393277,
     "end_time": "2020-05-08T03:34:42.657628",
     "exception": false,
     "start_time": "2020-05-08T03:34:35.264351",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count     12234574\n",
       "unique           6\n",
       "top           male\n",
       "freq       3645216\n",
       "Name: gender, dtype: object"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "book_gender = pd.read_csv(data_dir / 'author-gender.csv.gz')\n",
    "book_gender = book_gender.set_index('item')['gender']\n",
    "book_gender.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "papermill": {
     "duration": 5.524567,
     "end_time": "2020-05-08T03:34:48.236303",
     "exception": false,
     "start_time": "2020-05-08T03:34:42.711736",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[male, unlinked, female, unknown, ambiguous]\n",
       "Categories (5, object): [male, unlinked, female, unknown, ambiguous]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "book_gender[book_gender == 'no-viaf-author'] = 'unlinked'\n",
    "book_gender[book_gender == 'no-loc-author'] = 'unlinked'\n",
    "book_gender[book_gender == 'no-loc-book'] = 'unlinked'\n",
    "book_gender = book_gender.astype('category')\n",
    "book_gender.unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.057075,
     "end_time": "2020-05-08T03:34:48.350561",
     "exception": false,
     "start_time": "2020-05-08T03:34:48.293486",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "And we load book hashes, to set up our dummy bias:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "papermill": {
     "duration": 18.899345,
     "end_time": "2020-05-08T03:35:07.307688",
     "exception": false,
     "start_time": "2020-05-08T03:34:48.408343",
     "status": "completed"
    },
    "tags": []
   },
   "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>nisbns</th>\n",
       "      <th>md5</th>\n",
       "      <th>dcode</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>item</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>17</td>\n",
       "      <td>3781b82fabd530590c70cac955b52bb0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>4c6606ab43bfbe946a436c0ce7633a7a</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38</td>\n",
       "      <td>e16249d40bf94b35d8a784d73d0511c5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>289071ab1041c090ac252616a76fe079</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>7308735b39347b616ee6be0ab093541e</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      nisbns                               md5  dcode\n",
       "item                                                 \n",
       "0         17  3781b82fabd530590c70cac955b52bb0      0\n",
       "1          2  4c6606ab43bfbe946a436c0ce7633a7a      0\n",
       "2         38  e16249d40bf94b35d8a784d73d0511c5      1\n",
       "3          2  289071ab1041c090ac252616a76fe079      1\n",
       "4          4  7308735b39347b616ee6be0ab093541e      0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "book_hash = pd.read_parquet(data_dir / 'book-hash.parquet').rename(columns={'cluster': 'item'})\n",
    "book_hash['dcode'] = book_hash['md5'].apply(lambda x: int(x[-1], 16) % 2)\n",
    "book_hash = book_hash.set_index('item')\n",
    "book_hash.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.036231,
     "end_time": "2020-05-08T03:35:07.380090",
     "exception": false,
     "start_time": "2020-05-08T03:35:07.343859",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Load the sample user ratings for each data set:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "papermill": {
     "duration": 1.922147,
     "end_time": "2020-05-08T03:35:09.338882",
     "exception": false,
     "start_time": "2020-05-08T03:35:07.416735",
     "status": "completed"
    },
    "tags": []
   },
   "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>Set</th>\n",
       "      <th>user</th>\n",
       "      <th>item</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>nactions</th>\n",
       "      <th>first_time</th>\n",
       "      <th>last_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>462114</td>\n",
       "      <td>1.233101e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>1662785</td>\n",
       "      <td>1.233101e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>7287509</td>\n",
       "      <td>1.233101e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>8866889</td>\n",
       "      <td>1.233101e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>10031188</td>\n",
       "      <td>1.233101e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Set     user      item     timestamp  nactions  first_time  last_time\n",
       "0  AZ  4975592    462114  1.233101e+09       NaN         NaN        NaN\n",
       "1  AZ  4975592   1662785  1.233101e+09       NaN         NaN        NaN\n",
       "2  AZ  4975592   7287509  1.233101e+09       NaN         NaN        NaN\n",
       "3  AZ  4975592   8866889  1.233101e+09       NaN         NaN        NaN\n",
       "4  AZ  4975592  10031188  1.233101e+09       NaN         NaN        NaN"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_ratings = pd.read_csv(data_dir / 'study-ratings.csv')\n",
    "user_ratings.drop(columns=['rating'], inplace=True)\n",
    "user_ratings.rename(columns={'dataset': 'Set'}, inplace=True)\n",
    "user_ratings.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "papermill": {
     "duration": 6.717506,
     "end_time": "2020-05-08T03:35:16.116636",
     "exception": false,
     "start_time": "2020-05-08T03:35:09.399130",
     "status": "completed"
    },
    "tags": []
   },
   "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>Set</th>\n",
       "      <th>user</th>\n",
       "      <th>item</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>nactions</th>\n",
       "      <th>first_time</th>\n",
       "      <th>last_time</th>\n",
       "      <th>gender</th>\n",
       "      <th>dcode</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>462114</td>\n",
       "      <td>1.233101e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>1662785</td>\n",
       "      <td>1.233101e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>7287509</td>\n",
       "      <td>1.233101e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>8866889</td>\n",
       "      <td>1.233101e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>unknown</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>10031188</td>\n",
       "      <td>1.233101e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>43930</td>\n",
       "      <td>1.233274e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>11005013</td>\n",
       "      <td>1.233274e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>3649898</td>\n",
       "      <td>1.233619e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>8462465</td>\n",
       "      <td>1.233619e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>3727790</td>\n",
       "      <td>1.234656e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>856425</td>\n",
       "      <td>1.234829e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>4457778</td>\n",
       "      <td>1.234829e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>1551960</td>\n",
       "      <td>1.235002e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>2835913</td>\n",
       "      <td>1.238026e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>unknown</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>AZ</td>\n",
       "      <td>4975592</td>\n",
       "      <td>8845927</td>\n",
       "      <td>1.238544e+09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Set     user      item     timestamp  nactions  first_time  last_time  \\\n",
       "0   AZ  4975592    462114  1.233101e+09       NaN         NaN        NaN   \n",
       "1   AZ  4975592   1662785  1.233101e+09       NaN         NaN        NaN   \n",
       "2   AZ  4975592   7287509  1.233101e+09       NaN         NaN        NaN   \n",
       "3   AZ  4975592   8866889  1.233101e+09       NaN         NaN        NaN   \n",
       "4   AZ  4975592  10031188  1.233101e+09       NaN         NaN        NaN   \n",
       "5   AZ  4975592     43930  1.233274e+09       NaN         NaN        NaN   \n",
       "6   AZ  4975592  11005013  1.233274e+09       NaN         NaN        NaN   \n",
       "7   AZ  4975592   3649898  1.233619e+09       NaN         NaN        NaN   \n",
       "8   AZ  4975592   8462465  1.233619e+09       NaN         NaN        NaN   \n",
       "9   AZ  4975592   3727790  1.234656e+09       NaN         NaN        NaN   \n",
       "10  AZ  4975592    856425  1.234829e+09       NaN         NaN        NaN   \n",
       "11  AZ  4975592   4457778  1.234829e+09       NaN         NaN        NaN   \n",
       "12  AZ  4975592   1551960  1.235002e+09       NaN         NaN        NaN   \n",
       "13  AZ  4975592   2835913  1.238026e+09       NaN         NaN        NaN   \n",
       "14  AZ  4975592   8845927  1.238544e+09       NaN         NaN        NaN   \n",
       "\n",
       "     gender  dcode  \n",
       "0      male    0.0  \n",
       "1      male    1.0  \n",
       "2      male    1.0  \n",
       "3   unknown    0.0  \n",
       "4      male    0.0  \n",
       "5      male    1.0  \n",
       "6   unknown    1.0  \n",
       "7      male    1.0  \n",
       "8      male    0.0  \n",
       "9      male    0.0  \n",
       "10     male    0.0  \n",
       "11  unknown    1.0  \n",
       "12     male    0.0  \n",
       "13  unknown    0.0  \n",
       "14  unknown    1.0  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_ratings = user_ratings.join(book_gender, on='item', how='left')\n",
    "user_ratings['gender'].fillna('unlinked', inplace=True)\n",
    "user_ratings = user_ratings.join(book_hash['dcode'], on='item', how='left')\n",
    "user_ratings.head(15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.060949,
     "end_time": "2020-05-08T03:35:16.238148",
     "exception": false,
     "start_time": "2020-05-08T03:35:16.177199",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Now we will summarize user profiles:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "papermill": {
     "duration": 0.089575,
     "end_time": "2020-05-08T03:35:16.389783",
     "exception": false,
     "start_time": "2020-05-08T03:35:16.300208",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def summarize_profile(df):\n",
    "    gender = df['gender']\n",
    "    dc = df['dcode']\n",
    "    data = {\n",
    "        'count': len(df),\n",
    "        'linked': np.sum(gender != 'unlinked'),\n",
    "        'ambiguous': np.sum(gender == 'ambiguous'),\n",
    "        'male': np.sum(gender == 'male'),\n",
    "        'female': np.sum(gender == 'female'),\n",
    "        'dcknown': dc.count(),\n",
    "        'dcyes': dc.sum(skipna=True),\n",
    "        'PropDC': dc.mean()\n",
    "    }\n",
    "    data['Known'] = data['male'] + data['female']\n",
    "    data['PropFemale'] = data['female'] / data['Known']\n",
    "    data['PropKnown'] = data['Known'] / data['count']\n",
    "    return pd.Series(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "papermill": {
     "duration": 79.232407,
     "end_time": "2020-05-08T03:36:35.684635",
     "exception": false,
     "start_time": "2020-05-08T03:35:16.452228",
     "status": "completed"
    },
    "tags": []
   },
   "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>count</th>\n",
       "      <th>linked</th>\n",
       "      <th>ambiguous</th>\n",
       "      <th>male</th>\n",
       "      <th>female</th>\n",
       "      <th>dcknown</th>\n",
       "      <th>dcyes</th>\n",
       "      <th>PropDC</th>\n",
       "      <th>Known</th>\n",
       "      <th>PropFemale</th>\n",
       "      <th>PropKnown</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Set</th>\n",
       "      <th>user</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">AZ</th>\n",
       "      <th>529</th>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>5</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.625000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1723</th>\n",
       "      <td>25</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "      <td>6</td>\n",
       "      <td>25</td>\n",
       "      <td>14</td>\n",
       "      <td>0.560000</td>\n",
       "      <td>21</td>\n",
       "      <td>0.285714</td>\n",
       "      <td>0.840000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1810</th>\n",
       "      <td>14</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>0.125000</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.428571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2781</th>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>0.625000</td>\n",
       "      <td>6</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2863</th>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>6</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          count  linked  ambiguous  male  female  dcknown  dcyes    PropDC  \\\n",
       "Set user                                                                     \n",
       "AZ  529       8       8          2     1       4        8      3  0.375000   \n",
       "    1723     25      24          3    15       6       25     14  0.560000   \n",
       "    1810     14       6          0     6       0        8      1  0.125000   \n",
       "    2781      8       8          1     5       1        8      5  0.625000   \n",
       "    2863      6       6          0     6       0        6      4  0.666667   \n",
       "\n",
       "          Known  PropFemale  PropKnown  \n",
       "Set user                                \n",
       "AZ  529       5    0.800000   0.625000  \n",
       "    1723     21    0.285714   0.840000  \n",
       "    1810      6    0.000000   0.428571  \n",
       "    2781      6    0.166667   0.750000  \n",
       "    2863      6    0.000000   1.000000  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profiles = user_ratings.groupby(['Set', 'user']).apply(summarize_profile)\n",
    "profiles = profiles.apply(lambda s: s if s.name.startswith('Prop') else s.astype('i4'))\n",
    "profiles.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.061786,
     "end_time": "2020-05-08T03:36:35.809283",
     "exception": false,
     "start_time": "2020-05-08T03:36:35.747497",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "How are profile sizes distributed?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "papermill": {
     "duration": 2.877552,
     "end_time": "2020-05-08T03:36:38.750065",
     "exception": false,
     "start_time": "2020-05-08T03:36:35.872513",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x144 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.FacetGrid(profiles.reset_index(), col='Set', sharex=False, sharey=False, height=2)\n",
    "g.map(sns.distplot, 'count')\n",
    "plt.savefig(fig_dir / 'profile-size-all.pdf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "papermill": {
     "duration": 2.759299,
     "end_time": "2020-05-08T03:36:41.571838",
     "exception": false,
     "start_time": "2020-05-08T03:36:38.812539",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsgAAACICAYAAAD6bB0zAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAe+0lEQVR4nO3debxcZZ3n8c/3bgmQsGRhkS0IgRFsRYwIzdLSKgR0wLXBtgUcFF/zgh6XVhu08eXQ0C3NDDooOoMNLaZVcFTatAsoLQwgYQnKFmLgEgLGAElIzEa2e+s3f5ynQqWoe2/l3lP3nLr3+369zivnnq2ec+qXOr966nmeo4jAzMzMzMwyHUUXwMzMzMysTJwgm5mZmZnVcIJsZmZmZlbDCbKZmZmZWQ0nyGZmZmZmNZwgm5mZmZnVGPMJsqTPS1og6RFJD0l68xDbnyvpVTm+/iclbZK0W82yy1NZqtMTkvolTcrrdW1kioobSTMkbUyv+bCkeyQdlta9R9J/1Gx7fNquq8Fx7pC0qCbGfjDSstnwjZF4mjXS8tjwFXkvkzRT0k8kPSXpQUm3Szqx5nVWpDL9TtInBznOEkmP1nwuXZ1H+WzHjaF4mpZHmRqKiDE7AccC84AJ6e9pwKuG2OcOYFaOZbgfuAs4d5BtvgNcVvT18lR83AAzgMdq/v4YcEPN3z8F/hLoAh4B/rSV5fHkeMqzPJ7aMoYmAk8Ap9cse231ngacC3wtzU8FVgL7D3CsJcC0oq/neJ8cT81Nr6gpGGP2AVZGxGaAiFhZXSHpjcBVwCSyN+Bc4DhgFvAdSRuBYyNi43BfXNLB6fifAT4HfKvBNn8FHAKcM9zXsdwVGjd1dgVW1/z918BtwBHAAxFxT06vY63jeLKRKjKGPgjMi4i51QUR8RjwWP2GEfGipN5U3t8P8/Ws9RxPzSj6m0wrJ7I3+CGybytfB/4sLe8G7gGmp7/PBK6PIb4lkSW6DzWYrh5g+78DLiFryrIE2LNu/QxgKXBI0dfKUzniJsXExrT+KeA54IC6bf4RWMMg35xTeRbVvNaVRV/X8TqNoXhyDfL4jKGrgI8PUrZzebnG74B0nIkDbLsEeLTm9T5Z9LUdj9MYiyfXIA9HRKxP34ZOAE4CbpJ0ETCfrEr/l5IAOsluHEMd70rgyh0owlnAuyOiIulHwPuBawAkdQL/ClwSEb07cExrsRLEzVMRcSSApDOBa4HZ6e8O4G3AeuBAsm/4A/lgRMzfgde1FhhD8WQFKUEMbSPpZmAm8EREvCctPlPSScBhwEcjYtMghzgpamosbfSNsXhqmTGdIANERD/ZN587JD1K1pThQWBBRBy7I8eS9Bmynwfq3RkR/61u29eRvenVQOsBFpMSZLLa5eci4l92pAw2OoqKmwbmArUxcgHZT1GXANdIOjYiQtKtwF7A/Ij4yI6Uz1rP8WQjVWAMLQBOrCnHu1OHzf9Rs81NEXGhpGOBn0r6ObAilQ9gbkR8YUfKaK3leBramE6QU2/tSkQ8mRYdCTxD9tPz9HQzmCepGzg0IhYA64DJjY63g9+SPgB8MSL+saY8T0s6kKw9zbnAUcM4LWuxguOm3vFkP40jaW/gU8DREbFC0keBjwDfjIhThnl8azHHk41UwTH0XeBiSafHy+1Gdx7guPMkzSH7Cf3iVE4rGcdTc8Z0gkzWzuarknYH+oBe4PyI2CLpfcDVyoZf6wK+QvbN5lvA/86hIfpZwKl1y25Oy/+cLCBuT7XLVe+NiKeG+XqWnyLjBuBgSQ8BAraQJS2Qtd36p4hYkf7+BHCXpB9GxKoGx6l2qICsQ8bbRlAmG76xEk9WnMJiKCI2SnoncJWkrwAvkCVLlw2wyxXAbyT9Q0Ssa7D+dkn9af6RiDh7OOWyERlL8dQySg2dzczMzMyMcfCgEDMzMzOzHeEE2czMzMyshhNkMzMzM7MaTpDNzMzMzGqULkGePXt2AJ7ymcY1x5JjKS+OJcdRHhxHjqW8OJZaH0ulS5BXrvQDdiwfjiXLi2PJ8uA4srw4llqvdAmymZmZmVmRnCCbmZmZmdUofYL83fue5bv3PVt0MWwMcBxZHvyZZHlyLFke/LmUv9InyGZmZmZmo8kJspmZmZlZDSfIZmZmZmY1nCBb7iTNlrRIUq+kixqsnyDpprT+Pkkz0vKpkm6XtF7S1+r2eaOkR9M+V0vS6JyNmZmZjTdOkC1XkjqBa4BTgcOBD0g6vG6z84DVEXEI8GXgirR8E3AJ8OkGh/4GcD4wM02z8y+9mZmZmRNky9/RQG9ELI6ILcCNwBl125wB3JDmfwC8VZIiYkNE3E2WKG8jaR9g14iYFxEBfBt4V0vPwszMzMYtJ8iWt32B39f8vTQta7hNRPQBa4CpQxxz6RDHBEDS+ZLmS5q/YsWKHSy6mZmZmRNky1+jtsH1zzpvZpthbR8R10bErIiYNX369EEOaWZmZtaYE2TL21Jg/5q/9wOWDbSNpC5gN2DVEMfcb4hjmpmZmeXCCbLl7QFgpqSDJPUAZwFz67aZC5yT5t8H/Cq1LW4oIp4D1kk6Jo1ecTbw4/yLbmZmZgZdRRfAxpaI6JN0IXAr0AlcHxELJF0KzI+IucB1wBxJvWQ1x2dV95e0BNgV6JH0LuDkiHgc+K/At4CdgJ+nyczMzCx3TSXIkmYD/4ss4fnniPhS3foJZCMLvBF4ETgzIpZIejvwJaAH2AJ8JiJ+lWP5rYQi4mfAz+qWfaFmfhPw/gH2nTHA8vnAa/MrpZmZmVljQzaxGOG4tiuB/xwRf0L2k/qcvApuZmZmZtYKzbRBHsm4tr+NiGpnqgXAxFTbbGZmZmZWSs0kyHmNa/te4LcRsbn+BTx2rZmZmZmVRTMJ8ojHtZV0BFmzi481egGPXWtmZmZmZdFMgjyicW0l7QfcDJwdEU+NtMBmZmZmRZA0W9IiSb2SLmqwfoKkm9L6+yTNqFl3cVq+SNIpadn+km6XtFDSAkkfH72zscE0kyAPe1xbSbsDPwUujohf51VoMzMzs9E0kkEL0nZnAUcAs4Gvp+P1AX8TEa8BjgEuaHBMK8CQCXJqU1wd13Yh8P3quLaSTk+bXQdMTePafgqofqu6EDgEuETSQ2naM/ezMLMxKe/ampp1nZJ+K+knrT8LMxsjhj1oQVp+Y0RsjoingV7g6Ih4LiJ+AxAR68jyrPp+XlaApsZBHu64thFxGXDZCMtoZuNQTW3N28macT0gaW56cEzVttoaSWeR1dacWVdb8yrgNkmHRkR/2u/jZDeiXUfpdMys/TUatODNA22THpxVHbRgX+Deun23S4TTF/w3APc1enFJ5wPnAxxwwAHDPAVrlh81bWZllXttDWzrF/EO4J9H4RzMbOwYyaAFQw1mMAn4IfCJiFjb6MU9oMHocoJsZmU1kiEmB9v3K8BngcpgL+7hJ82szkgGLRhwX0ndZMnxdyLiRy0pue0wJ8hmVla519ZIeiewPCIeHOrFXVtjZnWGPWhBWn5W6jdxEDATuD/94nUdsDAirhqVs7CmOEE2s7JqRW3NccDpkpaQNdn4c0n/2orCm9nYMpJBCyJiAfB94HHgFuCC1CfiOOBDZJ9F1cEMThvVE7OGmuqkZ2ZWgG21NcAfyGpr/rJum2ptzTy2H2JyLvBdSVeRddKbCdwfEfOAiwEkvQX4dET81WicjJm1v+EOWpDWXQ5cXrfsbhr/4mUFc4JsZqWUeoBXa2s6geurtTXA/IiYS1ZbMyfV1qwiS6JJ21Vra/p4ubbGrFQqEUQE2S/tZlYWTpDNrLTyrq2pW38HcEce5TQbrr/7t8e4/+lVXP2BNxRdFDOr4TbIZmZmBahUsj6ncx+ub1pvZkVzgmxmZlaANRu3Fl0EMxuAE2QzM7MCvLhhc9FFMLMBOEE2MzMrwMr1WwDocP88s9JxgmxmZmOepNmSFknqlXRRg/UTJN2U1t8naUbNuovT8kWSTqlZvkTSo2ns2vk7WqYXU4K8ywT3lzcrG/+vNDOzMU1SJ3AN8Hayh8g8IGluRDxes9l5wOqIOETSWcAVwJmSDicbPvAIsjG1b5N0aM2wgSdFxMrhlKvaxGKSE2Sz0nENspmZjXVHA70RsTgitpA9RfGMum3OAG5I8z8A3poeA3wGcGNEbI6Ip4HedLwRW7Uhq0Heqbszj8OZWY6cIJuZ2Vi3L/D7mr+XpmUNt0mPFF4DTB1i3wB+IelBSec3emFJ50uaL2n+ihUrtlu3tb8CwJb0r5mVhxNkMzMb6xp1g4smtxls3+Mi4ijgVOACSSe+YsOIayNiVkTMmj59+nbrqnnxpq1OkM3Kxgmy5a6MnWHMbFxbCuxf8/d+QP3TObZtI6kL2I3s8eUD7hsR1X+XAzezg00v+itZYrxpq5+CblY2TpAtVzWdYU4FDgc+kDq51NrWGQb4MllnGOo6w8wGvp6OV3VSRBwZEbNafBpmNrY8AMyUdJCkHrLPmbl128wFzknz7wN+FRGRlp+VvtgfBMwE7pe0i6TJAJJ2AU4GHtuRQvWlJ+k5QTYrHyfIlrdSdoYxs/ErtSm+ELgVWAh8PyIWSLpU0ulps+uAqZJ6gU8BF6V9FwDfBx4HbgEuSCNY7AXcLelh4H7gpxFxy46Uqz8lyH2VoM/tkM1KxWPLWN4adWh580DbRESfpNrOMPfW7VvfGSaA/xMR17ag7GY2RkXEz4Cf1S37Qs38JuD9A+x7OXB53bLFwOtHUqZqDTLApr4KkzpdZ2VWFk39bxxum1JJUyXdLmm9pK/lW3QrqcI6w8DgPcbNzMqkv78mQXYzC7NSGTJBHkmbUmATcAnw6dxKbGVXaGeYwXqMm5mVyXY1yE6QzUqlmRrkYbcpjYgNEXE3WaJs40MpO8OYmZVNJZwgm5VVMwnySAZYb4p/Fh87ytoZxsysbGprkDdsdoLcDlo0jOn1kpZLcsVPiTSTII+kTWlT/LP42BIRP4uIQyPi4NS5hYj4QkTMTfObIuL9EXFIRBydOrtU97087XdYRPw8LVscEa9P0xHVY9rYl/fNSNL+qV/EQkkLJH189M7GbHvVcZABNmzuK7Ak1owWDmP6rbTMSqSZBHkkbUrNzIalRTejPuBvIuI1wDFkHT7rj2k2KvpqOumtc4LcDloyjGlE3IlzptJpJkEeSZtSM7Phyv1mFBHPRcRvACJiHVkzoPomY2ajor8STOzObsPrNzlBbgMjaXLazL6DcnPU0TVkgjySNqWQPSIYuAo4V9JS19aYWZNaejNKzTHeANzX6MV9M7JW66sEO/dkjyNY7xrkdtCqYUyb4uaoo6upB4WMcID1GSMon5mNXy27GUmaBPwQ+ERErG304ulhNNcCzJo1y7+IWe76K8FO3Vkz1HWbthZcGmvCjjQ5XdrsMKZWTn5sj5mVVUvG1JbUTZYcfyciftSSkps1oa9SobtTTOjqcBvk9pD7MKajVG4bBifIZlZWrRhTW2RNwhZGxFWjchZmA6hUoENi8sQut0FuAy0axhRJ3wPmAYelpqjnjeZ5WWNNNbEwMxttEdEnqXoz6gSur96MgPlp2MDrgDnpZrSKLIkmbVe9GfWRbkaSjgc+BDwq6aH0Up9LzcjMRlVfpUJHh5g0octtkNvECJucXg68YpjSiPhAzsW0HDhBNrPSyvtmlJ7s2ah9stmo668EHYLJE7tdg2xWMk6QzczMCtBXCTokNmzuo9ed9GyEnluzkYgga0lmI+U2yGZmZgXoTwnyLhO6/CQ9G5HfPruar/6ql9sXLS+6KGOGE2QzM7MC9KUmFrtO7GKtm1jYCCx8LhutctUG/xKRFyfIZmZmBeivBB0dYvLEbrb0VdxRz4Zt5fotgDtY5MkJspmZWQGqTSx23SnrDvTC2k0Fl8ja1drUht1fsvLjBNnMzKwA1QR58sRuwAmyDV+15thPZMyPE2QzM7MC9FUqaZi3rAb5+TVOkG14KpH96ycy5scJspmZWQGqbZCn7NxDh2Dxig1FF8naVF+lAsA6d/bMjRNkMzOzAlTHQe7q7GDKLhN4cvm6ootkbahSCbb2Z1XIfuBMfpwgm5mZFaC/P+hMjUf3nDyBJ5evL7ZA1pY291W2zbuTXn6cIJuZmRWgr/LyU8/23m0iS1Zu2DYagVmzNm7t3zbvTnr5cYJsZmZWgOooFgCvnrYLlYD7F68quFTWbrZPkF2DnBcnyGZmZgXoj6Aj3YX3n7IzXR3inqdeLLZQ1nY2pQR50oQuepevZ0tNkwsbPifIZmZmoywitqtB7u7s4MCpO3PPUysLLpm1m41bsgR55p6T6KsEC5atKbhEY4MTZDMzs1HWnwaurSbIAAdPn8Tvnl/Hsj9uLKpY1oaqNcgH7zkJgPlLVhdZnDGj9AlyfyV4fs0mtvb7JwMbmfWb+/jGHU9t+7ZtNlyrX9rCPb2u6bPh60sJcufL+TGv2293ero6uOKW3xVUKmtH1TbIU3buYcouPTywxO3Y89BVdAGG8uveldyy4Hn+uHELl57x2qKLY23sX379NM+t2cTUXXr4izftX3RxrE1VIrjy1kUALLx0Njv1dBZcImtH22qQO17OkKfs0sNxB0/jxw8tY8GytXz+tNdw0n/as6giWpuoVvp0d3UwY+rOzH9mNZv7+pnQ5c+mkWiqBlnSbEmLJPVKuqjB+gmSbkrr75M0o2bdxWn5Ikmn7GgBf7/6JQBue/wFImJHd7cCtCJehjrmULb2V1i+djMA8xa7E0y7KGMsPfPiS9vm73xyxQ6fkxWjbLH04vot1WNst/wth03nwKk707t8PR+b8yDX3N5Lrx8gUhpliyNg2/jZ3Z3idfvtzqoNW/jM/32EzX3+tXQkhqxBltQJXAO8HVgKPCBpbkQ8XrPZecDqiDhE0lnAFcCZkg4HzgKOAF4F3Cbp0Iho6l17fNlaFixbC8CyNZt4x9V3c8oRe/O6/XZj1526WLuxj9fvvzuTJ3bRKdHRIforQSWC7s7tc/9Kg2/rVRHxig+pdrNxSz8TuzsKP49WxEvaZ6hjDmrOvGfoj6BT4t8fXsZr992NNxywO/vtsRMTujqZ2N1BT2fj61f7xazo69tqff0V+iNKUfNQxljasLmP3zz7cvu+j815kHOOPZCD95zErAOnsM9uE+nqFF0dHelfbYuZrEMWbO0PerpK37ptTCljLJ0/Z37D5d2dHXz0hFezesMW5j68jCtvXcRVv3yCPzt0OgdN24V9d9+JDsFbX7MX0ydPYPVLW5g0oYtKwNqNW+mvBNMnT0DK2jf3dHY0vO+VTURQCbY1p+yvBBs297FszSbuXfwi7zpyX/bebWKhZSxjHC1fu2nbL1rdnR0cutdkTj58L+Y+vIy7e1fyjj/Zh9OPfBV/fGkr++6+E3vvNpGIYEt/hb13nTim72mVStCfOsNu2trPH/64kaWrN3Ly4Xs1dd7NNLE4GuiNiMUAkm4EzgBq37wzgC+m+R8AX1P26mcAN0bEZuBpSb3pePOaObnLfpq9xIkzp9PdJR54ehVfvu2JAbfv7BACguybVKWSPZ+8EtDVIfoqkSWRZBemen02bu2nu7ODzgYXrH5RGUMpgJe29LP3rhO5629PesWXg1HWinihiWMOaOX6zVz6k2zTj5xwELcueIG//8krd+1OiU1V9b3vrwR9lew/WU9XB9V7TX0ctbtKBJu2VpDgQ8ccWIYmTaWLpdsWvsCDz6xmyi49fPDNBzBn3jPcMO+ZQffp6hBdndp2bSNgQlcHXW2QtOThm2fP4k8PmVZ0MUoXSx8+bgZ/+8NH2dDgyWcdElMnTeDDxx3E6pe2cOuC57nziRXc+cSKbW2Xv/jvzdUP9HR20NkhgiAiu19Q82NsR8f2HQUhi9HB1H/mVbePdODa/aXss1LKPmMqkSUuHR2iU2Jrf4UgS5BjkNe+8f5nmfvXx7PrxO6hTrmVShdHe+46kZ6uDrb0VehJ9/63HLYn++2xMw8sWcV373+WOfc2/ozq6Wqc92TlGPg1Kw3eJDXIjnb03jicuKtsi5usorM2pqrlrD/uuX86gy+efsSQ5WkmQd4X+H3N30uBNw+0TUT0SVoDTE3L763bd9/6F5B0PnB++nO9pEU1q6fNAfeGacIzQM/nt/05DbglImaPcjFaFS9DHRMYOpY+d0XbxNI0Coz7v09TDcfSy7E07RlY+dsdO5eiFRZPx20fSEXEERQYS0N8JvFtmPbt9rnHFRZHzwC7fXa7Rf5Mqru/nd8+97dGRi22/nuaajSMpWYS5EbfAerz/IG2aWZfIuJa4NqGLy7Nj4hZQxXStpeuWxE3olbES6Mq8YbfNcdKLLVTWVuolLHUju9NO5Y5Z4XF0mCfSdBe7007lbVFSvmZBO3/3pSx/M38Fr8UqO3yvx+wbKBtJHUBuwGrmtzXxpZWxIvjaHxyLFleHEuWB8fRONJMgvwAMFPSQZJ6yBqZz63bZi5wTpp/H/CryHo2zQXOSr06DwJmAvfnU3QrqVbESzPHtLHHsWR5cSxZHhxH48iQTSxSG5oLgVuBTuD6iFgg6VJgfkTMBa4D5qRG56vI3mDSdt8na2zeB1zQ7AgWNQb8acoGVch1a1W8NDrmMIrXTrHUTmVtiRLHUju+N+1Y5tyUOJagvd6bdipr7hxHLVW68stjC5uZmZmZvcyDcZqZmZmZ1XCCbGZmZmZWo7QJskb46MWxRtL+km6XtFDSAkkfT8unSPqlpCfTv3uk5ZJ0dbp+j0g6quZY56Ttn5R0zkCvOVaUIZYGef++KOkPkh5K02k1+7TksaQ2fGW69pKWSHo0xc38tMyfB22iDLHk+8rYUIZYSuW4XtJySY/VLMstliS9MX3m9aZ9W/u0pYgo3UTWUP0p4NVAD/AwcHjR5Sr4muwDHJXmJwNPAIcD/wRclJZfBFyR5k8Dfk429uIxwH1p+RRgcfp3jzS/R9HnN9ZjaZD374vApxtsf3gq6wTgoHQOnWU5n/E4le3aA0uAaXXL/HnQBlNZYsn3lfafyhJLqSwnAkcBj9Usyy2WyEb9ODbt83Pg1FaeT1lrkLc9zjEitgDVRy+OWxHxXET8Js2vAxaSPYXnDOCGtNkNwLvS/BnAtyNzL7C7pH2AU4BfRsSqiFgN/BIo4oEio6UUsTTI+zeQbY8ljYingepjSUtxPuNUO1x7fx60h1LEku8rY0IpYgkgIu4kG7mjVi6xlNbtGhHzIsuWv11zrJYoa4Lc6HGOgyUT44qkGcAbgPuAvSLiOcg+7IA902YDXcPxdm1Ld7517x/AheknpuurPz/h96+MynbtA/iFpAeVPYIW/HnQLkp33X1faVtlv/Z5xdK+ab5+ecuUNUFu6hHV45GkScAPgU9ExNrBNm2wrOnHf48hpTrfBu/fN4CDgSOB54D/Wd20we7j8f0rk7Jd++Mi4ijgVOACSScOsq3jqVxKdd19X2lr7XrtdzSWRv08y5og+9GLDUjqJvsQ+05E/CgtfiH99ED6d3la7sdaZkpzvo3ev4h4ISL6I6ICfJPs5zLw+1dGpbr2EbEs/bscuJksdvx50B5Kc919X2l7Zb/2ecXS0jRfv7xlypog+9GLdVJvzeuAhRFxVc2q2sdangP8uGb52amn6DHAmvTzxq3AyZL2SD/nn5yWjVWliKWB3r/qB0fybqDa+9ePJS2f0lx7SbtImlydJ/t//Bj+PGgXpYgl31fGhFLE0iByiaW0bp2kY1Lcnl1zrNZoZQ/AkUxkPRyfIOud+fmiy1P0BBxP9nPCI8BDaToNmAr8B/Bk+ndK2l7ANen6PQrMqjnWfyHr9NULfLjocxsPsTTI+zcnvT+PkH1g7FOzz+dTmRdR01u3DOczXqeyXHuyHusPp2lBtSz+PGifqQyx5PvK2JjKEEupHN8jayq4lazG97w8YwmYRVYR8BTwNdLToFs1+VHTZmZmZmY1ytrEwszMzMysEE6QzczMzMxqOEE2MzMzM6vhBNnMzMzMrIYTZDMzMzOzGuM2QZa0vmb+NElPSjqgyDJZe3IsWR4cR5YXx5LlZTzHUlfRBSiapLcCXwVOjohniy6PtS/HkuXBcWR5cSxZXsZjLI3bGmQASSeQPd73HRHxVFr2LUlXS7pH0mJJ70vLJelKSY9JelTSmWn51yWdnuZvlnR9mj9P0mWSZkhaKOmbkhZI+oWknYo5Y2sVx5LlwXFkeXEsWV7GayyN5wR5AtljCt8VEb+rW7cP2ROG3gl8KS17D3Ak8HrgbcCVyh4TfCdwQtpmX+DwNH88cFeanwlcExFHAH8E3pv72ViRHEuWB8eR5cWxZHkZt7E0nhPkrcA9ZI9CrPdvEVGJiMeBvdKy44HvRUR/RLwA/D/gTWRv7AmSDgceB15IwXBsOj7A0xHxUJp/EJjRihOywjiWLA+OI8uLY8nyMm5jaTwnyBXgL4A3Sfpc3brNNfOq+3c7EfEHYA9gNtk3pLvScddHxLoGx+vHbb/HGseS5cFxZHlxLFlexm0sjecEmYh4ieyngQ9KavTtqNadwJmSOiVNB04E7k/r5gGf4OU3/dO8/JOBjQOOJcuD48jy4liyvIzXWBr33/QiYpWk2cCdklYOsunNZD8FPAwE8NmIeD6tu4usZ2evpGeAKZT4TbfWcCxZHhxHlhfHkuVlPMaSIqLoMpiZmZmZlca4bmJhZmZmZlbPCbKZmZmZWQ0nyGZmZmZmNZwgm5mZmZnVcIJsZmZmZlbDCbKZmZmZWQ0nyGZmZmZmNf4/uDL92Qy4ysIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x144 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.FacetGrid(profiles.reset_index(), col='Set', sharex=False, sharey=False, height=2)\n",
    "g.map(sns.distplot, 'Known')\n",
    "plt.savefig(fig_dir / 'profile-size-known.pdf')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.065854,
     "end_time": "2020-05-08T03:36:41.701767",
     "exception": false,
     "start_time": "2020-05-08T03:36:41.635913",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "For the paper, we want to make some changes - we're going to show this on a scatter plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "papermill": {
     "duration": 0.143,
     "end_time": "2020-05-08T03:36:41.907513",
     "exception": false,
     "start_time": "2020-05-08T03:36:41.764513",
     "status": "completed"
    },
    "tags": []
   },
   "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>Set</th>\n",
       "      <th>user</th>\n",
       "      <th>Type</th>\n",
       "      <th>Size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AZ</td>\n",
       "      <td>529</td>\n",
       "      <td>All</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AZ</td>\n",
       "      <td>1723</td>\n",
       "      <td>All</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AZ</td>\n",
       "      <td>1810</td>\n",
       "      <td>All</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AZ</td>\n",
       "      <td>2781</td>\n",
       "      <td>All</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AZ</td>\n",
       "      <td>2863</td>\n",
       "      <td>All</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Set  user Type  Size\n",
       "0  AZ   529  All     8\n",
       "1  AZ  1723  All    25\n",
       "2  AZ  1810  All    14\n",
       "3  AZ  2781  All     8\n",
       "4  AZ  2863  All     6"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "up_sizes = profiles[['count', 'Known']].reset_index().melt(id_vars=['Set', 'user'], var_name='Type', value_name='Size')\n",
    "up_sizes['Type'] = up_sizes['Type'].astype('category').cat.rename_categories({\n",
    "    'count': 'All',\n",
    "    'Known': 'Known-Gender'\n",
    "})\n",
    "up_sizes.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "papermill": {
     "duration": 0.207822,
     "end_time": "2020-05-08T03:36:42.154324",
     "exception": false,
     "start_time": "2020-05-08T03:36:41.946502",
     "status": "completed"
    },
    "tags": []
   },
   "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>Set</th>\n",
       "      <th>Type</th>\n",
       "      <th>Size</th>\n",
       "      <th>Users</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AZ</td>\n",
       "      <td>Known-Gender</td>\n",
       "      <td>5</td>\n",
       "      <td>1300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AZ</td>\n",
       "      <td>Known-Gender</td>\n",
       "      <td>6</td>\n",
       "      <td>829</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AZ</td>\n",
       "      <td>Known-Gender</td>\n",
       "      <td>7</td>\n",
       "      <td>572</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AZ</td>\n",
       "      <td>Known-Gender</td>\n",
       "      <td>8</td>\n",
       "      <td>416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AZ</td>\n",
       "      <td>Known-Gender</td>\n",
       "      <td>9</td>\n",
       "      <td>284</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Set          Type  Size  Users\n",
       "0  AZ  Known-Gender     5   1300\n",
       "1  AZ  Known-Gender     6    829\n",
       "2  AZ  Known-Gender     7    572\n",
       "3  AZ  Known-Gender     8    416\n",
       "4  AZ  Known-Gender     9    284"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "size_counts = up_sizes.groupby(['Set', 'Type', 'Size'])['user'].count().reset_index(name='Users')\n",
    "size_counts = size_counts[size_counts['Users'] > 0]\n",
    "size_counts.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "papermill": {
     "duration": 0.099594,
     "end_time": "2020-05-08T03:36:42.324672",
     "exception": false,
     "start_time": "2020-05-08T03:36:42.225078",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    4202.000000\n",
       "mean       11.899096\n",
       "std        46.498697\n",
       "min         1.000000\n",
       "25%         1.000000\n",
       "50%         2.000000\n",
       "75%         7.000000\n",
       "max      1300.000000\n",
       "Name: Users, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "size_counts['Users'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "papermill": {
     "duration": 17.040578,
     "end_time": "2020-05-08T03:36:59.435841",
     "exception": false,
     "start_time": "2020-05-08T03:36:42.395263",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/MICHAELEKSTRAND/anaconda3/envs/bookfair/lib/python3.7/site-packages/plotnine/ggplot.py:729: PlotnineWarning: Saving 7 x 3.2 in image.\n",
      "  from_inches(height, units), units), PlotnineWarning)\n",
      "/home/MICHAELEKSTRAND/anaconda3/envs/bookfair/lib/python3.7/site-packages/plotnine/ggplot.py:730: PlotnineWarning: Filename: figures/ProfileData/profile-size.pdf\n",
      "  warn('Filename: {}'.format(filename), PlotnineWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8782675280353)>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "make_plot(size_counts, aes(x='Size', y='Users'),\n",
    "          geom_point(),\n",
    "          scale_x_log10(),\n",
    "          scale_y_log10(),\n",
    "          facet_grid('Type ~ Set', scales='free'),\n",
    "          xlab('# of Consumed Items'),\n",
    "          ylab('# of Users'),\n",
    "          panel_grid=element_blank(),\n",
    "          file='profile-size.pdf', width=7, height=3.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.073588,
     "end_time": "2020-05-08T03:36:59.582513",
     "exception": false,
     "start_time": "2020-05-08T03:36:59.508925",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "And % known?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "papermill": {
     "duration": 2.38235,
     "end_time": "2020-05-08T03:37:02.029726",
     "exception": false,
     "start_time": "2020-05-08T03:36:59.647376",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7fce06753bd0>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x216 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.FacetGrid(profiles.reset_index(), col='Set')\n",
    "g.map(sns.distplot, 'PropKnown')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.078553,
     "end_time": "2020-05-08T03:37:02.187238",
     "exception": false,
     "start_time": "2020-05-08T03:37:02.108685",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Distribution of Female Authors\n",
    "\n",
    "Quick empirical inspection:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "papermill": {
     "duration": 1.906982,
     "end_time": "2020-05-08T03:37:04.173942",
     "exception": false,
     "start_time": "2020-05-08T03:37:02.266960",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7fce067169d0>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x216 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.FacetGrid(profiles.reset_index(), col='Set', sharey=False)\n",
    "g.map(sns.distplot, 'PropFemale')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "papermill": {
     "duration": 0.111915,
     "end_time": "2020-05-08T03:37:04.368749",
     "exception": false,
     "start_time": "2020-05-08T03:37:04.256834",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Set\n",
       "AZ      0.414445\n",
       "BX-E    0.418886\n",
       "BX-I    0.407030\n",
       "GR-E    0.446998\n",
       "GR-I    0.450201\n",
       "Name: PropFemale, dtype: float64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profiles.groupby('Set')['PropFemale'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "papermill": {
     "duration": 0.112704,
     "end_time": "2020-05-08T03:37:04.559536",
     "exception": false,
     "start_time": "2020-05-08T03:37:04.446832",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Set\n",
       "AZ      0.329436\n",
       "BX-E    0.267354\n",
       "BX-I    0.254304\n",
       "GR-E    0.276480\n",
       "GR-I    0.269127\n",
       "Name: PropFemale, dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sqrt(profiles.groupby('Set')['PropFemale'].var())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.085467,
     "end_time": "2020-05-08T03:37:04.731342",
     "exception": false,
     "start_time": "2020-05-08T03:37:04.645875",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Distribution of Dummy Codes\n",
    "\n",
    "Quick empirical inspection - this should be noise:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "papermill": {
     "duration": 3.167559,
     "end_time": "2020-05-08T03:37:07.984538",
     "exception": false,
     "start_time": "2020-05-08T03:37:04.816979",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7fcdd3aebc10>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x1080 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.FacetGrid(profiles.reset_index(), row='Set', sharey=False)\n",
    "g.map(sns.distplot, 'PropDC')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.089866,
     "end_time": "2020-05-08T03:37:08.161719",
     "exception": false,
     "start_time": "2020-05-08T03:37:08.071853",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Saving Outputs\n",
    "\n",
    "We save profile data frame to be reloaded in the Bayesian analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "papermill": {
     "duration": 0.115371,
     "end_time": "2020-05-08T03:37:08.367180",
     "exception": false,
     "start_time": "2020-05-08T03:37:08.251809",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "profiles.to_pickle('data/profile-data.pkl')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.089919,
     "end_time": "2020-05-08T03:37:08.548234",
     "exception": false,
     "start_time": "2020-05-08T03:37:08.458315",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "We also want to save the data for STAN."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "papermill": {
     "duration": 0.10957,
     "end_time": "2020-05-08T03:37:08.754619",
     "exception": false,
     "start_time": "2020-05-08T03:37:08.645049",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def stan_inputs(data, kc, pc):\n",
    "    return {\n",
    "        'J': len(data),\n",
    "        'n': data[kc],\n",
    "        'y': data[pc]\n",
    "    }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "papermill": {
     "duration": 0.105411,
     "end_time": "2020-05-08T03:37:08.950211",
     "exception": false,
     "start_time": "2020-05-08T03:37:08.844800",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def inf_dir(sname):\n",
    "    return data_dir / sname / 'inference'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "papermill": {
     "duration": 0.153444,
     "end_time": "2020-05-08T03:37:09.195378",
     "exception": false,
     "start_time": "2020-05-08T03:37:09.041934",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "preparing STAN input for AZ\n",
      "preparing STAN input for BX-E\n",
      "preparing STAN input for BX-I\n",
      "preparing STAN input for GR-E\n",
      "preparing STAN input for GR-I\n"
     ]
    }
   ],
   "source": [
    "for sname, frame in profiles.groupby('Set'):\n",
    "    print('preparing STAN input for', sname)\n",
    "    dir = inf_dir(sname)\n",
    "    dir.mkdir(exist_ok=True)\n",
    "    in_fn = dir / 'profile-inputs.json'\n",
    "    in_fn.write_text(ujson.dumps(stan_inputs(frame, 'Known', 'female')))\n",
    "    in_fn = dir / 'profile-dcode-inputs.json'\n",
    "    in_fn.write_text(ujson.dumps(stan_inputs(frame, 'dcknown', 'dcyes')))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.6"
  },
  "papermill": {
   "duration": 159.623951,
   "end_time": "2020-05-08T03:37:10.892003",
   "environment_variables": {},
   "exception": null,
   "input_path": "ProfileDataPrep.ipynb",
   "output_path": "ProfileDataPrep.temp.ipynb",
   "parameters": {},
   "start_time": "2020-05-08T03:34:31.268052",
   "version": "1.0.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}