analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import os\n",
    "from sklearn.metrics import f1_score\n",
    "from matplotlib import pyplot as plt\n",
    "import matplotlib.colors as mcolors\n",
    "from pathlib import Path\n",
    "\n",
    "from IPython.display import display\n",
    "\n",
    "\n",
    "os.environ['CUDA_LAUNCH_BLOCKING'] = '1'\n",
    "\n",
    "import statsmodels.formula.api as smf\n",
    "import statsmodels.api as sm\n",
    "import statsmodels.stats.anova as anova\n",
    "from statsmodels.stats.multicomp import pairwise_tukeyhsd\n",
    "from statsmodels.graphics.factorplots import interaction_plot\n",
    "from statsmodels.formula.api import ols\n",
    "from statsmodels.stats.anova import anova_lm\n",
    "from statsmodels.stats.multicomp import pairwise_tukeyhsd\n",
    "from statsmodels.sandbox.stats.multicomp import MultiComparison\n",
    "\n",
    "from const import MASK_TOKEN\n",
    "\n",
    "pd.options.display.float_format = '{:.4f}'.format"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "BATCH_SIZE = 64\n",
    "DEVICE = 'cuda'\n",
    "LR = 1e-5\n",
    "NUM_ITER = 30\n",
    "\n",
    "out_dir_base = Path('data/exp_access/reliance_model')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import load_exp_results\n",
    "import importlib\n",
    "importlib.reload(load_exp_results)\n",
    "valid_user, task_result, f_score, questionnaire, open_question_answer = load_exp_results.load()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "age 48.358620689655176 13.117352725635445 19 81\n",
      "     crowd_id  register_time  submit_password  age  got_consent  \\\n",
      "sex                                                               \n",
      "1          46             46               46   46           46   \n",
      "2          98             98               98   98           98   \n",
      "3           1              1                1    1            1   \n",
      "\n",
      "     is_no_more_login  task_mode  python_class  \n",
      "sex                                             \n",
      "1                  46         46            46  \n",
      "2                  98         98            98  \n",
      "3                   1          1             1  \n",
      "                            f_score  accuracy  rate  reliance\n",
      "condition                                                    \n",
      "RandomCueUser_Proportional       50        50    50        50\n",
      "RandomCueUser_Threshold          95        95    95        95\n",
      "SelectiveCueUser                 50        50    50        50\n",
      "SelectiveCueUser2                52        52    52        52\n",
      "SelectiveCueUser3                98        98    98        98\n",
      "age 47.357142857142854 13.682142775451329 17 86\n",
      "     crowd_id  register_time  submit_password  age  got_consent  \\\n",
      "sex                                                               \n",
      "1          38             38               38   38           38   \n",
      "2          55             55               55   55           55   \n",
      "3           5              5                5    5            5   \n",
      "\n",
      "     is_no_more_login  task_mode  python_class  \n",
      "sex                                             \n",
      "1                  38         38            38  \n",
      "2                  55         55            55  \n",
      "3                   5          5             5  \n"
     ]
    }
   ],
   "source": [
    "# DEMOGRAPHICS\n",
    "\n",
    "_user, _instruction_quiz_answer, _main_task_result, _questionnaire, _open_question_answer = load_exp_results._load_json()\n",
    "\n",
    "_user = _user[_user.index.isin(f_score.index)]\n",
    "_target = _user[_user['python_class'].isin(['RandomCueUser_Threshold', 'RandomCueUser_Proportional'])]\n",
    "# Reliance dataset\n",
    "print('age', _target['age'].mean(), _target['age'].std(), _target['age'].min(), _target['age'].max())\n",
    "print(_target.groupby('sex').count())\n",
    "print(f_score.groupby('condition').count())\n",
    "\n",
    "# Evaluation\n",
    "_target = _user[_user['python_class'].isin(['SelectiveCueUser3'])]\n",
    "print('age', _target['age'].mean(), _target['age'].std(), _target['age'].min(), _target['age'].max())\n",
    "print(_target.groupby('sex').count())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_1706447/3276330542.py:10: UserWarning: The palette list has more values (10) than needed (2), which may not be intended.\n",
      "  sns.swarmplot(data=_target, x='rate', y='f_score', hue='condition', hue_order=['RandomCueUser_Threshold','RandomCueUser_Proportional',], ax=ax, palette=sns.color_palette())\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                        df       F  PR(>F)  eta_partial\n",
      "C(condition)        1.0000  1.6467  0.2015       0.0115\n",
      "rate                1.0000 12.8807  0.0005       0.0837\n",
      "rate:C(condition)   1.0000  1.3755  0.2428       0.0097\n",
      "Residual          141.0000     NaN     NaN       0.5000\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 727x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "[{'value': 'ロボットの事をあまり考えずに答えてしまった'},\n",
       " {'value': 'ロボットのメッセージの表示が遅く感じた'},\n",
       " {'value': 'ロボットに任せるような、難しい内容ではなかった'},\n",
       " {'value': '認識が難しいものがあればロボットを使用するつもりでいたが、60問すべて比較的簡単に認識できたので、ロボットに任せる必要性を特に感じない問題でした。'},\n",
       " {'value': 'ロボットに任せたのは1回だけ。その回答があまりにかけ離れていたのでそれ以降は自分で回答した。また、自分で判読するに十分な「見えにくさ」だったと思う。\\r\\n実験の役に立てたのか疑問。'},\n",
       " {'value': '最初の数問、判読しにくい設問の時に、試しにロボットに入力させてみたら、全く的外れな回答だった。２回試してみてロボットは当てにならないと判断し、あとは全て人力で回答した。'},\n",
       " {'value': 'はじめはロボットはわざと間違った答えを出していると思った。\\r\\n当てにならないと後半思った'},\n",
       " {'value': '読みづらいと感じた時には役に立つ'},\n",
       " {'value': '一度読めるかを確認後入力の選択を実施したのでロボットに任せたのは2件だけだった。\\u3000読み辛かった2件ともロボットは不正解だったような気がするので画像認識は得意ではないと判断した。'},\n",
       " {'value': ''},\n",
       " {'value': 'ロボットに任せるような、難しい内容ではなかった'},\n",
       " {'value': '大文字、小文字の使い分けが不要なので、自分での入力が簡単にできました。そのうえで、判読が難しいものはロボットに任せたところ、出来が悪く思えたので、以後は自分の入力に努めました。'},\n",
       " {'value': '最初の不明瞭な文字をロボットに任せましたが、明らかにおかしな判定だったので、その後は自分で判断しました。'},\n",
       " {'value': 'ロボットの回答はあまり信頼できなかった'},\n",
       " {'value': 'ロボットの回答が全然信用できなかったので自力で回答したほうが断然よかった。最初にロボットが間違えると使いたくなる'},\n",
       " {'value': '今回は検査の意図をあまり理解しておらず、全て自身で入力してしまいました'},\n",
       " {'value': 'ロボットは発言がなかった\\r\\n早く回答するために自信がない問題でも自分で対応した'},\n",
       " {'value': '戦略を考える余裕なく作業に取り組んでいました'},\n",
       " {'value': '私の苦手は ロボットの苦手かも'}]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Figure 5\n",
    "\n",
    "sns.set()\n",
    "sns.set_theme(style=\"ticks\")\n",
    "sns.set_context(\"paper\")\n",
    "\n",
    "_target = f_score[(f_score['condition'] == 'RandomCueUser_Threshold') +  (f_score['condition'] == 'RandomCueUser_Proportional')]\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "sns.swarmplot(data=_target, x='rate', y='f_score', hue='condition', hue_order=['RandomCueUser_Threshold','RandomCueUser_Proportional',], ax=ax, palette=sns.color_palette())\n",
    "\n",
    "ax.set_ylabel('F-score')\n",
    "ax.set_ylim([-0.05, 1.05])\n",
    "handles, labels = ax.get_legend_handles_labels()\n",
    "ax.legend(handles, ['Threshold', 'Proportional'], loc='upper left')\n",
    "\n",
    "plt.savefig('dataset.pdf', bbox_inches='tight')\n",
    "plt.show()\n",
    "\n",
    "\n",
    "formula = 'f_score ~ rate * C(condition)'\n",
    "model = ols(formula, _target).fit()\n",
    "aov_table = anova_lm(model, typ=2)\n",
    "aov_table['eta_partial'] = aov_table['sum_sq'] / (aov_table['sum_sq'] + aov_table.loc['Residual', 'sum_sq'])\n",
    "aov_table = aov_table.drop('sum_sq', axis=1)\n",
    "print(aov_table)\n",
    "\n",
    "\n",
    "plt.clf()\n",
    "sns.lmplot(data=_target, x='rate', y='f_score', hue='condition')\n",
    "plt.show()\n",
    "\n",
    "\n",
    "_open_question_answer[_open_question_answer['user_id'].isin(\n",
    "    _target.query('rate<=0.2 and reliance < 0.1').index\n",
    ")]['answers'].tolist()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f_score      18\n",
      "accuracy     18\n",
      "condition    18\n",
      "rate         18\n",
      "reliance     18\n",
      "dtype: int64 f_score      13\n",
      "accuracy     13\n",
      "condition    13\n",
      "rate         13\n",
      "reliance     13\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Out-of-range data\n",
    "_target = f_score[ (f_score['condition'] == 'SelectiveCueUser3')]\n",
    "print(_target.query('rate <= 0.2').count(), _target.query('rate > 0.8').count())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "condition\n",
      "RandomCueUser_Threshold    47\n",
      "SelectiveCueUser3          67\n",
      "dtype: int64\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/seaborn/regression.py:582: UserWarning: legend_out is deprecated from the `lmplot` function signature. Please update your code to pass it using `facet_kws`.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 520x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f_score ~ rate * C(condition)\n"
     ]
    },
    {
     "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>sum_sq</th>\n",
       "      <th>df</th>\n",
       "      <th>F</th>\n",
       "      <th>PR(&gt;F)</th>\n",
       "      <th>eta_partial</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>C(condition)</th>\n",
       "      <td>0.3040</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>6.5748</td>\n",
       "      <td>0.0117</td>\n",
       "      <td>0.0564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rate</th>\n",
       "      <td>0.3528</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>7.6293</td>\n",
       "      <td>0.0067</td>\n",
       "      <td>0.0649</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rate:C(condition)</th>\n",
       "      <td>0.0641</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>1.3866</td>\n",
       "      <td>0.2415</td>\n",
       "      <td>0.0124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Residual</th>\n",
       "      <td>5.0867</td>\n",
       "      <td>110.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.5000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   sum_sq       df      F  PR(>F)  eta_partial\n",
       "C(condition)       0.3040   1.0000 6.5748  0.0117       0.0564\n",
       "rate               0.3528   1.0000 7.6293  0.0067       0.0649\n",
       "rate:C(condition)  0.0641   1.0000 1.3866  0.2415       0.0124\n",
       "Residual           5.0867 110.0000    NaN     NaN       0.5000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Figure 7\n",
    "_target = f_score[(f_score['condition'] == 'RandomCueUser_Threshold') +  (f_score['condition'] == 'SelectiveCueUser3')]\n",
    "\n",
    "_target = _target[(_target['rate'] > 0.2) * (_target['rate'] <= 0.8)]\n",
    "print(_target.groupby('condition').apply(len))\n",
    "palette=list(sns.color_palette())[3::-3]\n",
    "g = sns.lmplot(data=_target, x='rate', y='f_score', hue='condition', hue_order=['SelectiveCueUser3','RandomCueUser_Threshold', ], markers=['x', '+'], scatter_kws={'alpha':0.6},\n",
    "                x_ci='ci', palette=palette, legend_out=False, height=4, aspect=1.3)\n",
    "g.set_ylabels('F-score')\n",
    "g._legend.set_title(None)\n",
    "new_labels = ['vPred-RC', 'Random',]\n",
    "for t, l in zip(g._legend.texts, new_labels):\n",
    "    t.set_text(l)\n",
    "plt.savefig(\"f_score.pdf\", bbox_inches='tight')\n",
    "plt.show(g)\n",
    "\n",
    "\n",
    "formula = 'f_score ~ rate * C(condition)'\n",
    "model = ols(formula, _target).fit()\n",
    "aov_table = anova_lm(model, typ=2)\n",
    "aov_table['eta_partial'] = aov_table['sum_sq'] / (aov_table['sum_sq'] + aov_table.loc['Residual', 'sum_sq'])\n",
    "print(formula)\n",
    "display(aov_table)\n",
    " \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\subfigure[Instruction 1]{\n",
      "\\begin{tabular}{r|p{1.5em}|p{3.3em}|p{3.3em}|p{3.3em}}\n",
      "& df & F & PR(>F) & $\\eta_p^2$ \\\\ \\hline \n",
      "condition & 1.0000 & 0.0009 & 0.9761 & 0.0000\\\\ \n",
      "rate & 1.0000 & 0.9486 & 0.3322 & 0.0086\\\\ \n",
      "rate:condition & 1.0000 & 0.0031 & 0.9559 & 0.0000\\\\ \n",
      "Residual & 110.0000 & nan & nan & 0.5000\\\\ \n",
      "\n",
      "\\end{tabular}\n",
      "}\n",
      "\n",
      "\\subfigure[Instruction 2]{\n",
      "\\begin{tabular}{r|p{1.5em}|p{3.3em}|p{3.3em}|p{3.3em}}\n",
      "& df & F & PR(>F) & $\\eta_p^2$ \\\\ \\hline \n",
      "condition & 1.0000 & 0.0005 & 0.9827 & 0.0000\\\\ \n",
      "rate & 1.0000 & 0.5008 & 0.4807 & 0.0045\\\\ \n",
      "rate:condition & 1.0000 & 1.0116 & 0.3167 & 0.0091\\\\ \n",
      "Residual & 110.0000 & nan & nan & 0.5000\\\\ \n",
      "\n",
      "\\end{tabular}\n",
      "}\n",
      "\n",
      "\\subfigure[Performance understanding]{\n",
      "\\begin{tabular}{r|p{1.5em}|p{3.3em}|p{3.3em}|p{3.3em}}\n",
      "& df & F & PR(>F) & $\\eta_p^2$ \\\\ \\hline \n",
      "condition & 1.0000 & 1.4315 & 0.2341 & 0.0128\\\\ \n",
      "rate & 1.0000 & 0.0000 & 0.9944 & 0.0000\\\\ \n",
      "rate:condition & 1.0000 & 0.1279 & 0.7213 & 0.0012\\\\ \n",
      "Residual & 110.0000 & nan & nan & 0.5000\\\\ \n",
      "\n",
      "\\end{tabular}\n",
      "}\n",
      "\n",
      "\\subfigure[Performance predictability]{\n",
      "\\begin{tabular}{r|p{1.5em}|p{3.3em}|p{3.3em}|p{3.3em}}\n",
      "& df & F & PR(>F) & $\\eta_p^2$ \\\\ \\hline \n",
      "condition & 1.0000 & 0.0346 & 0.8528 & 0.0003\\\\ \n",
      "rate & 1.0000 & 1.2988 & 0.2569 & 0.0117\\\\ \n",
      "rate:condition & 1.0000 & 0.1215 & 0.7281 & 0.0011\\\\ \n",
      "Residual & 110.0000 & nan & nan & 0.5000\\\\ \n",
      "\n",
      "\\end{tabular}\n",
      "}\n",
      "\n",
      "\\subfigure[RCC useful for assignment*]{\n",
      "\\begin{tabular}{r|p{1.5em}|p{3.3em}|p{3.3em}|p{3.3em}}\n",
      "& df & F & PR(>F) & $\\eta_p^2$ \\\\ \\hline \n",
      "condition & 1.0000 & 3.9477 & 0.0494 & 0.0346\\\\ \n",
      "rate & 1.0000 & 5.5109 & 0.0207 & 0.0477\\\\ \n",
      "rate:condition & 1.0000 & 0.0956 & 0.7577 & 0.0009\\\\ \n",
      "Residual & 110.0000 & nan & nan & 0.5000\\\\ \n",
      "\n",
      "\\end{tabular}\n",
      "}\n",
      "\n",
      "\\subfigure[RCC useful for understanding performance*]{\n",
      "\\begin{tabular}{r|p{1.5em}|p{3.3em}|p{3.3em}|p{3.3em}}\n",
      "& df & F & PR(>F) & $\\eta_p^2$ \\\\ \\hline \n",
      "condition & 1.0000 & 4.2278 & 0.0421 & 0.0370\\\\ \n",
      "rate & 1.0000 & 3.3975 & 0.0680 & 0.0300\\\\ \n",
      "rate:condition & 1.0000 & 0.3265 & 0.5689 & 0.0030\\\\ \n",
      "Residual & 110.0000 & nan & nan & 0.5000\\\\ \n",
      "\n",
      "\\end{tabular}\n",
      "}\n",
      "\n",
      "\\subfigure[AI useful†]{\n",
      "\\begin{tabular}{r|p{1.5em}|p{3.3em}|p{3.3em}|p{3.3em}}\n",
      "& df & F & PR(>F) & $\\eta_p^2$ \\\\ \\hline \n",
      "condition & 1.0000 & 3.0444 & 0.0838 & 0.0269\\\\ \n",
      "rate & 1.0000 & 5.0823 & 0.0262 & 0.0442\\\\ \n",
      "rate:condition & 1.0000 & 0.8474 & 0.3593 & 0.0076\\\\ \n",
      "Residual & 110.0000 & nan & nan & 0.5000\\\\ \n",
      "\n",
      "\\end{tabular}\n",
      "}\n",
      "\n",
      "\\subfigure[AI trustable*]{\n",
      "\\begin{tabular}{r|p{1.5em}|p{3.3em}|p{3.3em}|p{3.3em}}\n",
      "& df & F & PR(>F) & $\\eta_p^2$ \\\\ \\hline \n",
      "condition & 1.0000 & 6.0794 & 0.0152 & 0.0524\\\\ \n",
      "rate & 1.0000 & 0.9756 & 0.3255 & 0.0088\\\\ \n",
      "rate:condition & 1.0000 & 0.7297 & 0.3949 & 0.0066\\\\ \n",
      "Residual & 110.0000 & nan & nan & 0.5000\\\\ \n",
      "\n",
      "\\end{tabular}\n",
      "}\n",
      "\n",
      "\\subfigure[Want to use again†]{\n",
      "\\begin{tabular}{r|p{1.5em}|p{3.3em}|p{3.3em}|p{3.3em}}\n",
      "& df & F & PR(>F) & $\\eta_p^2$ \\\\ \\hline \n",
      "condition & 1.0000 & 3.7007 & 0.0570 & 0.0325\\\\ \n",
      "rate & 1.0000 & 1.9920 & 0.1610 & 0.0178\\\\ \n",
      "rate:condition & 1.0000 & 0.4197 & 0.5185 & 0.0038\\\\ \n",
      "Residual & 110.0000 & nan & nan & 0.5000\\\\ \n",
      "\n",
      "\\end{tabular}\n",
      "}\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/root/.local/share/virtualenvs/kabu_summarizer-Xgg0Bd0D/lib/python3.8/site-packages/seaborn/regression.py:582: UserWarning: legend_out is deprecated from the `lmplot` function signature. Please update your code to pass it using `facet_kws`.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1713.25x600 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import re\n",
    "pd.options.display.float_format = '{:.4f}'.format\n",
    "\n",
    "q_labels = ['instruction1', 'instruction2', 'understandable', 'predictable', 'instance_conf_decision', 'instance_conf_understand', 'useful', 'trust', 'future']\n",
    "q_labels_new = ['Instruction 1', 'Instruction 2', 'Performance understanding', 'Performance predictability', 'RCC useful for assignment*', 'RCC useful for understanding performance*', 'AI useful†', 'AI trustable*', 'Want to use again†']\n",
    "q_label_dict = {l: l_new for (l, l_new) in zip(q_labels, q_labels_new)}\n",
    "\n",
    "questionnaire['condition'] = questionnaire.index.map(lambda uid: f_score.loc[uid, 'condition'])\n",
    "questionnaire['rate'] = questionnaire.index.map(lambda uid: f_score.loc[uid, 'rate'])\n",
    "questionnaire_target = questionnaire[(questionnaire['condition'] == 'RandomCueUser_Threshold') + (questionnaire['condition'] == 'SelectiveCueUser3')]\n",
    "questionnaire_target = questionnaire_target[(0.2 < questionnaire_target['rate']) * (0.8 >= questionnaire_target['rate'])]\n",
    "\n",
    "def to_latex(anova_table):\n",
    "    def _float_to_str(val):\n",
    "        return \"{:.4f}\".format(val)\n",
    "    columns = anova_table.columns\n",
    "    index = aov_table.index\n",
    "    index = map(lambda s: re.sub('C\\((.+)\\)', r'\\1', s), index)\n",
    "    values = anova_table.values.tolist()\n",
    "\n",
    "    ret = '& ' + ' & '.join(columns) + ' \\\\\\\\ \\\\hline \\n'\n",
    "    for index, line in zip(index, values):\n",
    "        ret += f'{index} & ' + ' & '.join(map(_float_to_str, line)) + '\\\\\\\\ \\n'\n",
    "    return ret\n",
    "    \n",
    "for column in ['instruction1', 'instruction2', 'understandable', 'predictable', 'instance_conf_decision', 'instance_conf_understand', 'useful', 'trust', 'future']:\n",
    "    if column in ['condition', 'rate']:\n",
    "        continue\n",
    "    _data = questionnaire_target[[column, 'condition', 'rate']]\n",
    "    label = q_label_dict[column]\n",
    "    print(f'\\\\subfigure[{label}]' + '{')\n",
    "    print(r'\\begin{tabular}{r|p{1.5em}|p{3.3em}|p{3.3em}|p{3.3em}}')  \n",
    "    formula = f'{column} ~ rate * C(condition)'\n",
    "    model = ols(formula, _data).fit()\n",
    "    aov_table = anova_lm(model, typ=2)\n",
    "    aov_table['$\\\\eta_p^2$'] = aov_table['sum_sq'] / (aov_table['sum_sq'] + aov_table.loc['Residual', 'sum_sq'])\n",
    "    aov_table = aov_table.drop('sum_sq', axis=1)\n",
    "    print(to_latex(aov_table))\n",
    "    print(r'\\end{tabular}')\n",
    "    print('}\\n')\n",
    "\n",
    "_target = questionnaire_target.reset_index().set_index(['user_id', 'rate', 'condition']).stack().reset_index().rename(columns={'item_name': 'question', 0: 'value'})\n",
    "_target = _target.query('value != 4.5')\n",
    "\n",
    "\n",
    "\n",
    "row = {label: r for (label, r) in zip(q_labels, [0,0,0,0,1,1,1,1,1])}\n",
    "column = {label: c for (label, c) in zip(q_labels, [0,1,2,3,0,1,2,3,4])}\n",
    "coordinates_rev = {(row[label], column[label]): label for label in q_labels}\n",
    "_target['row'] = _target['question'].map(row)\n",
    "_target['column'] = _target['question'].map(column)\n",
    "\n",
    "g = sns.lmplot(data=_target, col='column', row='row',\n",
    "               x='rate', y='value', hue='condition', hue_order=['SelectiveCueUser3','RandomCueUser_Threshold', ], \n",
    "               legend_out=True, palette=list(sns.color_palette())[3::-3], markers=['x', '+'], scatter_kws={'alpha':0.4},\n",
    "               height=3, )\n",
    "new_labels = ['vPred-RC', 'Random',]\n",
    "for t, l in zip(g._legend.texts, new_labels):\n",
    "    t.set_text(l)\n",
    "g._legend.set_title(None)\n",
    "sns.move_legend(g, \"upper left\", bbox_to_anchor=(.75, .8))\n",
    "\n",
    "for r, axes in enumerate(g.axes):\n",
    "    for c, ax in enumerate(axes):\n",
    "        ax.set_ylim([0.9, 5.1])\n",
    "        ax.set_yticks(range(1,6))\n",
    "        \n",
    "        if (r,c) in coordinates_rev.keys():\n",
    "            ax.set_title(q_label_dict[coordinates_rev[(r, c)]])\n",
    "        else:\n",
    "            ax.set_visible(False)\n",
    "\n",
    "\n",
    "plt.savefig('questionnaire.pdf', bbox_inches='tight')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import numpy as np\n",
    "\n",
    "import inference\n",
    "inference.init()\n",
    "\n",
    "def which_dtype(array):\n",
    "    if type(array) == torch.Tensor:\n",
    "        return array.dtype\n",
    "    if np.issubdtype(array.dtype, np.integer):\n",
    "        return torch.long\n",
    "    elif np.issubdtype(array.dtype, np.floating):\n",
    "        return torch.float32\n",
    "    elif np.issubdtype(array.dtype, np.bool_):\n",
    "        return torch.bool\n",
    "\n",
    "def get_input(user_id, target_index):\n",
    "    example = task_result[task_result['user_id'] == user_id].sort_values('episode_id')\n",
    "\n",
    "    src = dict()\n",
    "    src['middles'] = np.vstack(example.apply(lambda raw: inference.get_y(raw['dataset_name'], 'captcha-09az+capital-color', raw['ground_truth'], ['middle'])[0], axis=1).values).astype(np.float32)\n",
    "    src['cues'] = example['system_action_token'].values.copy()\n",
    "    src['decisions'] = example['user_decision_token'].values.copy()\n",
    "\n",
    "    src = {key: torch.tensor(value, dtype=which_dtype(value)) for key, value in src.items()}\n",
    "    mask = {'middles': torch.zeros((src['middles'].shape[0]), dtype=torch.bool)}\n",
    "\n",
    "    mask['middles'][target_index+1:] = True\n",
    "\n",
    "    ys = example.apply(lambda raw: inference.get_y(raw['dataset_name'], 'captcha-09az+capital-color', raw['ground_truth'], ['y'])[0], axis=1).values[target_index]\n",
    "    return src, mask, ys\n",
    "\n",
    "class Dataset(torch.utils.data.Dataset):\n",
    "    def __init__(self, user_ids, rtn_acc=False):\n",
    "        self.user_ids = user_ids\n",
    "        self.n_episode = 60\n",
    "        self.rtn_acc = rtn_acc\n",
    "        return\n",
    "\n",
    "    def __len__(self):\n",
    "        return self.n_episode * len(self.user_ids)\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        user_id = self.user_ids[idx // self.n_episode]\n",
    "        episode_idx = idx % self.n_episode\n",
    "        src, mask, ys = get_input(user_id, episode_idx)\n",
    "        length = src['middles'].shape[0]\n",
    "\n",
    "        src = {key: self._padding(value, self.n_episode, 0) for key, value in src.items()}\n",
    "        mask = {key: self._padding(value, self.n_episode, True) for key, value in mask.items()}\n",
    "\n",
    "        if length <= episode_idx:\n",
    "            # ignore_index\n",
    "            label = [torch.tensor(-100)]\n",
    "        else:\n",
    "            # 0: AI, 1: SELF\n",
    "            label = src['decisions'][episode_idx].clone() - 1\n",
    "        # add [MASK] token\n",
    "        src['decisions'][episode_idx:] = MASK_TOKEN\n",
    "\n",
    "        if self.rtn_acc:\n",
    "            conf = inference.instance_confidence_calculators['captcha-09az+capital-color'].calc_score(\n",
    "                ys.reshape((1, 36, 5))\n",
    "            )\n",
    "            acc = inference.acc_estimators['captcha-09az+capital-color'].predict_proba(conf.reshape((1, 1)))[0,1]\n",
    "            return (src, mask, idx % self.n_episode), label, acc\n",
    "        return (src, mask, idx % self.n_episode), label\n",
    "\n",
    "    def _padding(self, array, length, value=None):\n",
    "        orig_length = array.shape[0]\n",
    "        if orig_length == length:\n",
    "            return array\n",
    "        elif orig_length > length:\n",
    "            raise ValueError('Input array too long.')\n",
    "        addition = torch.concat([torch.empty_like(array[:1])] * (length - array.shape[0]))\n",
    "        if value is not None:\n",
    "            addition[:] = value\n",
    "        return torch.concat([array, addition])\n",
    "\n",
    "import trust_model\n",
    "models = list()\n",
    "for model_dir in out_dir_base.glob('[0-9]'):\n",
    "    model = trust_model.SimpleTransformerEncoder_Access(n_head=16, n_feature=128, dropout=0.2, n_hidden=2048, n_layers=3, n_out=2)\n",
    "    model.load_state_dict(torch.load(model_dir / '19.pth'))\n",
    "    model.eval()\n",
    "    model = model.to(DEVICE)\n",
    "    models.append(model)\n",
    "\n",
    "\n",
    "# hist of threshold (all)\n",
    "ids = pd.DataFrame(sum(\n",
    "[[{'name': f'{line[\"condition\"]}-{line[\"rate\"]}', 'index': item} for item in line[0]] \n",
    "    for _, line in pd.DataFrame(f_score.groupby(['condition', 'rate']).apply(lambda grp: grp.index)).reset_index().iterrows()],\n",
    "    list())).set_index('index')\n",
    "dataset = Dataset(ids.index, rtn_acc=True)\n",
    "dataloader = torch.utils.data.DataLoader(dataset, batch_size=1)\n",
    "diff_distribution = list()\n",
    "for (src, mask, target_idx), label, acc in dataloader:\n",
    "    #print(src['middles'].permute((1,0,2)).shape, src['cues'].permute((1,0)).shape)\n",
    "    src['middles'] = src['middles'].permute((1,0,2)).tile((1, 3, 1))\n",
    "    src['cues'] = src['cues'].permute((1,0)).tile((1, 3))\n",
    "    src['decisions'] = src['decisions'].permute((1,0)).tile((1, 3))\n",
    "    src['middles_mask'] = mask['middles'].tile(3, 1)\n",
    "    del mask\n",
    "    src['cues'][target_idx] = torch.tensor([1,2,3], dtype=torch.long)\n",
    "    for key, value in src.items():\n",
    "        src[key] = value.to(DEVICE)\n",
    "    p_current = torch.stack([model(**src)[target_idx[0]].softmax(1) for model in models]).mean(dim=0)[:, 0]\n",
    "    diff = (p_current - acc.item()).abs()\n",
    "    diff_distribution.append((diff[2] - diff[:2].min()).item())\n",
    "diff_distribution = np.array(diff_distribution)\n",
    "sns.histplot(diff_distribution)\n",
    "np.save('diff_distribution.npy', diff_distribution)\n",
    "\n",
    "\n",
    "\n",
    "diff_distribution = np.load('diff_distribution.npy')\n",
    "sns.histplot(diff_distribution)\n",
    "plt.savefig('diff_distribution.pdf')\n",
    "\n",
    "n = diff_distribution.shape[0]\n",
    "diff_distribution.sort()\n",
    "[diff_distribution[int(n * r)] for r in np.arange(0.2, 1, 0.2)][::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.23420780897140503,\n",
       " 0.14439991116523743,\n",
       " 0.08581850677728653,\n",
       " 0.049771592020988464,\n",
       " 0.026333104819059372]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Figure 6\n",
    "dataset = Dataset(f_score[(f_score['condition'] == 'RandomCueUser_Threshold') + (f_score['condition'] == 'RandomCueUser_Proportional')].index, rtn_acc=True)\n",
    "dataloader = torch.utils.data.DataLoader(dataset, batch_size=1)\n",
    "diff_distribution = list()\n",
    "for (src, mask, target_idx), label, acc in dataloader:\n",
    "    src['middles'] = src['middles'].permute((1,0,2)).tile((1, 3, 1))\n",
    "    src['cues'] = src['cues'].permute((1,0)).tile((1, 3))\n",
    "    src['decisions'] = src['decisions'].permute((1,0)).tile((1, 3))\n",
    "    src['middles_mask'] = mask['middles'].tile(3, 1)\n",
    "    del mask\n",
    "    src['cues'][target_idx] = torch.tensor([1,2,3], dtype=torch.long)\n",
    "    for key, value in src.items():\n",
    "        src[key] = value.to(DEVICE)\n",
    "    p_current = torch.stack([model(**src)[target_idx[0]].softmax(1) for model in models]).mean(dim=0)[:, 0]\n",
    "    diff = (p_current - acc.item()).abs()\n",
    "    diff_distribution.append((diff[2] - diff[:2].min()).item())\n",
    "diff_distribution = np.array(diff_distribution)\n",
    "sns.histplot(diff_distribution)\n",
    "np.save('diff_distribution_test_random.npy', diff_distribution)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(6,2))\n",
    "\n",
    "diff_distribution = np.load('diff_distribution_test_random.npy')\n",
    "sns.histplot(diff_distribution, color=sns.color_palette()[0], bins=300)\n",
    "plt.savefig('diff_distribution_test_random.pdf', bbox_inches='tight')\n",
    "\n",
    "\n",
    "n = diff_distribution.shape[0]\n",
    "diff_distribution.sort()\n",
    "[diff_distribution[int(n * r)] for r in np.arange(0.3, 0.8, 0.1)][::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_2828062/53022933.py:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  _target['rate'] = _target['user_id'].map(lambda user_id: f_score.loc[user_id, 'rate'])\n",
      "/tmp/ipykernel_2828062/53022933.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  _target['f_score'] = _target['user_id'].map(lambda user_id: f_score.loc[user_id, 'f_score'])\n"
     ]
    }
   ],
   "source": [
    "_target = open_question_answer[open_question_answer['user_id'].isin(\n",
    "                     f_score[(f_score['condition'] == 'SelectiveCueUser3') * (f_score['rate'] > 0.2) * (f_score['rate'] <= 0.8)].index\n",
    ")]\n",
    "_target['rate'] = _target['user_id'].map(lambda user_id: f_score.loc[user_id, 'rate'])\n",
    "_target['f_score'] = _target['user_id'].map(lambda user_id: f_score.loc[user_id, 'f_score'])\n",
    "_target = _target.drop(columns=['register_time', 'task_name', 'item_name'])\n",
    "import csv\n",
    "\n",
    "with open('open_question_answer.csv', 'w', encoding='utf-8') as f:\n",
    "    writer = csv.writer(f)\n",
    "    writer.writerow(_target.columns.values.tolist())\n",
    "    for line in _target.values.tolist():\n",
    "        writer.writerow(line)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_2828062/3618408028.py:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  _target['rate'] = _target['user_id'].map(lambda user_id: f_score.loc[user_id, 'rate'])\n",
      "/tmp/ipykernel_2828062/3618408028.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  _target['f_score'] = _target['user_id'].map(lambda user_id: f_score.loc[user_id, 'f_score'])\n"
     ]
    }
   ],
   "source": [
    "_target = open_question_answer[open_question_answer['user_id'].isin(\n",
    "                     f_score[(f_score['condition'] == 'RandomCueUser_Threshold') * (f_score['rate'] > 0.2) * (f_score['rate'] <= 0.8)].index\n",
    ")]\n",
    "_target['rate'] = _target['user_id'].map(lambda user_id: f_score.loc[user_id, 'rate'])\n",
    "_target['f_score'] = _target['user_id'].map(lambda user_id: f_score.loc[user_id, 'f_score'])\n",
    "_target = _target.drop(columns=['register_time', 'task_name', 'item_name'])\n",
    "import csv\n",
    "\n",
    "with open('open_question_answer.csv', 'w', encoding='utf-8') as f:\n",
    "    writer = csv.writer(f)\n",
    "    writer.writerow(_target.columns.values.tolist())\n",
    "    for line in _target.values.tolist():\n",
    "        writer.writerow(line)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_3147771/4212318241.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  _target['target_rate'] = cue_rates[_target.index % 5]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "PearsonRResult(statistic=0.6786678051223041, pvalue=1.6035473994896456e-14)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Correlation between threshold and rate\n",
    "import numpy as np\n",
    "cue_rates = np.arange(0.3, 0.8, 0.1)\n",
    "_target = f_score[ (f_score['condition'] == 'SelectiveCueUser3')]\n",
    "\n",
    "_target['target_rate'] = cue_rates[_target.index % 5]\n",
    "\n",
    "sns.lmplot(data=_target, x='target_rate', y='rate')\n",
    "from scipy.stats import pearsonr\n",
    "pearsonr(_target['target_rate'], _target['rate'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_3147771/2237917444.py:4: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  performance = task_result[task_result['user_class'] == 'SelectiveCueUser3'].query(\"episode_id < 6\").sort_values('episode_id').groupby('user_id').mean()\n"
     ]
    },
    {
     "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>episode_id</th>\n",
       "      <th>ai_correct</th>\n",
       "      <th>final_answer_correct</th>\n",
       "      <th>cue_shown</th>\n",
       "      <th>system_action_token</th>\n",
       "      <th>user_decision_token</th>\n",
       "      <th>assignment_correct</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>user_id</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>409</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>413</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.1667</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>1.1667</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>423</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>1.3333</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>424</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.3333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>427</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>2.3333</td>\n",
       "      <td>1.1667</td>\n",
       "      <td>1.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>441</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>2.3333</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>444</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.3333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>464</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.3333</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>468</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.1667</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>1.1667</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>469</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>1.3333</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>478</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>503</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.3333</td>\n",
       "      <td>0.3333</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>513</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>514</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>1.3333</td>\n",
       "      <td>1.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>527</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.3333</td>\n",
       "      <td>2.3333</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>528</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>531</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>2.0000</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>534</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>1.3333</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>539</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>545</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.3333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.3333</td>\n",
       "      <td>2.6667</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>550</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.1667</td>\n",
       "      <td>2.6667</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>552</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>0.8333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>557</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>2.0000</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>1.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>567</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.1667</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>2.1667</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>1.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>572</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>1.3333</td>\n",
       "      <td>1.5000</td>\n",
       "      <td>1.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>578</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>2.0000</td>\n",
       "      <td>1.3333</td>\n",
       "      <td>1.0000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         episode_id  ai_correct  final_answer_correct  cue_shown  \\\n",
       "user_id                                                            \n",
       "409          2.5000      0.5000                0.8333     0.8333   \n",
       "413          2.5000      0.1667                0.8333     1.0000   \n",
       "423          2.5000      0.5000                0.8333     1.0000   \n",
       "424          2.5000      0.3333                0.8333     0.8333   \n",
       "427          2.5000      0.8333                1.0000     0.5000   \n",
       "441          2.5000      0.5000                1.0000     0.5000   \n",
       "444          2.5000      0.3333                0.8333     0.6667   \n",
       "464          2.5000      0.3333                0.6667     0.8333   \n",
       "468          2.5000      0.1667                0.8333     1.0000   \n",
       "469          2.5000      0.5000                0.8333     1.0000   \n",
       "478          2.5000      0.8333                0.8333     0.6667   \n",
       "503          2.5000      0.3333                0.3333     0.6667   \n",
       "513          2.5000      0.6667                0.8333     0.8333   \n",
       "514          2.5000      0.6667                0.8333     0.8333   \n",
       "527          2.5000      0.8333                0.8333     0.3333   \n",
       "528          2.5000      0.5000                0.8333     0.8333   \n",
       "531          2.5000      0.5000                0.8333     0.6667   \n",
       "534          2.5000      0.5000                1.0000     1.0000   \n",
       "539          2.5000      0.8333                0.8333     0.6667   \n",
       "545          2.5000      0.3333                0.8333     0.3333   \n",
       "550          2.5000      0.8333                0.8333     0.1667   \n",
       "552          2.5000      0.5000                1.0000     0.6667   \n",
       "557          2.5000      0.5000                0.5000     0.5000   \n",
       "567          2.5000      0.1667                0.8333     0.5000   \n",
       "572          2.5000      0.5000                0.8333     1.0000   \n",
       "578          2.5000      0.6667                0.8333     0.6667   \n",
       "\n",
       "         system_action_token  user_decision_token  assignment_correct  \n",
       "user_id                                                                \n",
       "409                   1.5000               1.6667              0.8333  \n",
       "413                   1.1667               1.6667              0.8333  \n",
       "423                   1.3333               1.6667              0.8333  \n",
       "424                   1.5000               1.8333              0.8333  \n",
       "427                   2.3333               1.1667              1.0000  \n",
       "441                   2.3333               1.6667              0.8333  \n",
       "444                   1.6667               1.5000              0.8333  \n",
       "464                   1.5000               1.8333              0.8333  \n",
       "468                   1.1667               1.6667              0.8333  \n",
       "469                   1.6667               1.3333              0.8333  \n",
       "478                   1.8333               1.0000              0.8333  \n",
       "503                   1.6667               1.5000              0.8333  \n",
       "513                   1.8333               1.5000              0.8333  \n",
       "514                   1.6667               1.3333              1.0000  \n",
       "527                   2.3333               1.0000              0.8333  \n",
       "528                   1.5000               1.6667              0.8333  \n",
       "531                   2.0000               1.6667              0.8333  \n",
       "534                   1.3333               1.6667              0.8333  \n",
       "539                   1.8333               1.0000              0.8333  \n",
       "545                   2.6667               1.5000              0.8333  \n",
       "550                   2.6667               1.0000              0.8333  \n",
       "552                   1.8333               1.6667              0.8333  \n",
       "557                   2.0000               1.5000              1.0000  \n",
       "567                   2.1667               1.8333              1.0000  \n",
       "572                   1.3333               1.5000              1.0000  \n",
       "578                   2.0000               1.3333              1.0000  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>episode_id</th>\n",
       "      <th>ai_correct</th>\n",
       "      <th>final_answer_correct</th>\n",
       "      <th>cue_shown</th>\n",
       "      <th>system_action_token</th>\n",
       "      <th>user_decision_token</th>\n",
       "      <th>assignment_correct</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>user_id</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>445</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.1667</td>\n",
       "      <td>2.8333</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>0.1667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>463</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.3333</td>\n",
       "      <td>2.6667</td>\n",
       "      <td>2.0000</td>\n",
       "      <td>0.1667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>498</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>0.1667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>546</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>2.5000</td>\n",
       "      <td>1.8333</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>565</th>\n",
       "      <td>2.5000</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.8333</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>2.3333</td>\n",
       "      <td>1.6667</td>\n",
       "      <td>0.1667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         episode_id  ai_correct  final_answer_correct  cue_shown  \\\n",
       "user_id                                                            \n",
       "445          2.5000      0.6667                0.5000     0.1667   \n",
       "463          2.5000      0.8333                1.0000     0.3333   \n",
       "498          2.5000      0.6667                0.8333     0.8333   \n",
       "546          2.5000      0.8333                0.8333     0.5000   \n",
       "565          2.5000      0.8333                0.8333     0.5000   \n",
       "\n",
       "         system_action_token  user_decision_token  assignment_correct  \n",
       "user_id                                                                \n",
       "445                   2.8333               1.8333              0.1667  \n",
       "463                   2.6667               2.0000              0.1667  \n",
       "498                   1.8333               1.8333              0.1667  \n",
       "546                   2.5000               1.8333              0.0000  \n",
       "565                   2.3333               1.6667              0.1667  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.set_option('display.max_rows', 100)\n",
    "#display(task_result)\n",
    "task_result['assignment_correct'] = task_result['ai_correct']== (task_result['user_decision'] == 'AI')\n",
    "performance = task_result[task_result['user_class'] == 'SelectiveCueUser3'].query(\"episode_id < 6\").sort_values('episode_id').groupby('user_id').mean()\n",
    "display(performance.query('assignment_correct >= 5.0/6'))\n",
    "display(performance.query('assignment_correct <= 1.0/6'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "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>user_id</th>\n",
       "      <th>episode_id</th>\n",
       "      <th>dataset_name</th>\n",
       "      <th>ground_truth</th>\n",
       "      <th>ai_answer</th>\n",
       "      <th>potential_cue</th>\n",
       "      <th>user_decision</th>\n",
       "      <th>system_action</th>\n",
       "      <th>final_answer</th>\n",
       "      <th>user_class</th>\n",
       "      <th>ai_correct</th>\n",
       "      <th>final_answer_correct</th>\n",
       "      <th>cue_shown</th>\n",
       "      <th>assignment_correct</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</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",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>24512</th>\n",
       "      <td>545</td>\n",
       "      <td>0</td>\n",
       "      <td>captcha-images</td>\n",
       "      <td>rhzrx</td>\n",
       "      <td>e7c2x</td>\n",
       "      <td>unconfident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>neutral</td>\n",
       "      <td>rhzrx</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24527</th>\n",
       "      <td>545</td>\n",
       "      <td>1</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>khcys</td>\n",
       "      <td>khcys</td>\n",
       "      <td>confident</td>\n",
       "      <td>AI</td>\n",
       "      <td>confident</td>\n",
       "      <td>khcys</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24554</th>\n",
       "      <td>545</td>\n",
       "      <td>2</td>\n",
       "      <td>captcha-images</td>\n",
       "      <td>qngsz</td>\n",
       "      <td>xmme2</td>\n",
       "      <td>unconfident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>neutral</td>\n",
       "      <td>qngsz</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24564</th>\n",
       "      <td>545</td>\n",
       "      <td>3</td>\n",
       "      <td>captcha-09az</td>\n",
       "      <td>wbncw</td>\n",
       "      <td>wbccw</td>\n",
       "      <td>confident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>neutral</td>\n",
       "      <td>wbncw</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24574</th>\n",
       "      <td>545</td>\n",
       "      <td>4</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>wfe4c</td>\n",
       "      <td>wfe4c</td>\n",
       "      <td>confident</td>\n",
       "      <td>AI</td>\n",
       "      <td>confident</td>\n",
       "      <td>wfe4c</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24579</th>\n",
       "      <td>545</td>\n",
       "      <td>5</td>\n",
       "      <td>captcha-images</td>\n",
       "      <td>rbiol</td>\n",
       "      <td>f7fx2</td>\n",
       "      <td>unconfident</td>\n",
       "      <td>AI</td>\n",
       "      <td>neutral</td>\n",
       "      <td>f7fx2</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       user_id  episode_id    dataset_name ground_truth ai_answer  \\\n",
       "id                                                                  \n",
       "24512      545           0  captcha-images        rhzrx     e7c2x   \n",
       "24527      545           1   capital-color        khcys     khcys   \n",
       "24554      545           2  captcha-images        qngsz     xmme2   \n",
       "24564      545           3    captcha-09az        wbncw     wbccw   \n",
       "24574      545           4   capital-color        wfe4c     wfe4c   \n",
       "24579      545           5  captcha-images        rbiol     f7fx2   \n",
       "\n",
       "      potential_cue user_decision system_action final_answer  \\\n",
       "id                                                             \n",
       "24512   unconfident          SELF       neutral        rhzrx   \n",
       "24527     confident            AI     confident        khcys   \n",
       "24554   unconfident          SELF       neutral        qngsz   \n",
       "24564     confident          SELF       neutral        wbncw   \n",
       "24574     confident            AI     confident        wfe4c   \n",
       "24579   unconfident            AI       neutral        f7fx2   \n",
       "\n",
       "              user_class  ai_correct  final_answer_correct  cue_shown  \\\n",
       "id                                                                      \n",
       "24512  SelectiveCueUser3       False                  True      False   \n",
       "24527  SelectiveCueUser3        True                  True       True   \n",
       "24554  SelectiveCueUser3       False                  True      False   \n",
       "24564  SelectiveCueUser3       False                  True      False   \n",
       "24574  SelectiveCueUser3        True                  True       True   \n",
       "24579  SelectiveCueUser3       False                 False      False   \n",
       "\n",
       "       assignment_correct  \n",
       "id                         \n",
       "24512                True  \n",
       "24527                True  \n",
       "24554                True  \n",
       "24564                True  \n",
       "24574                True  \n",
       "24579               False  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Figure 8 (a) (successful example)\n",
    "_target = task_result[task_result['user_class'] == 'SelectiveCueUser3']\n",
    "columns = ['user_id', 'episode_id', 'dataset_name', 'ground_truth', 'ai_answer',\n",
    "       'potential_cue', 'user_decision', 'system_action',\n",
    "       'final_answer', 'user_class', 'ai_correct', 'final_answer_correct', 'cue_shown',\n",
    "       'assignment_correct']\n",
    "display(_target.query(\"episode_id < 6\").sort_values('episode_id').query('user_id == 545')[columns])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>user_id</th>\n",
       "      <th>episode_id</th>\n",
       "      <th>dataset_name</th>\n",
       "      <th>ground_truth</th>\n",
       "      <th>ai_answer</th>\n",
       "      <th>potential_cue</th>\n",
       "      <th>user_decision</th>\n",
       "      <th>system_action</th>\n",
       "      <th>final_answer</th>\n",
       "      <th>user_class</th>\n",
       "      <th>ai_correct</th>\n",
       "      <th>final_answer_correct</th>\n",
       "      <th>cue_shown</th>\n",
       "      <th>assignment_correct</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</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",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>25655</th>\n",
       "      <td>565</td>\n",
       "      <td>0</td>\n",
       "      <td>captcha-09az</td>\n",
       "      <td>fg38b</td>\n",
       "      <td>fg38b</td>\n",
       "      <td>confident</td>\n",
       "      <td>AI</td>\n",
       "      <td>neutral</td>\n",
       "      <td>fg38b</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25677</th>\n",
       "      <td>565</td>\n",
       "      <td>1</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>nrby8</td>\n",
       "      <td>nrby8</td>\n",
       "      <td>confident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>neutral</td>\n",
       "      <td>nrby8</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25706</th>\n",
       "      <td>565</td>\n",
       "      <td>2</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>s3ych</td>\n",
       "      <td>s3ych</td>\n",
       "      <td>unconfident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>unconfident</td>\n",
       "      <td>s3ych</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25714</th>\n",
       "      <td>565</td>\n",
       "      <td>3</td>\n",
       "      <td>captcha-09az</td>\n",
       "      <td>5bnd7</td>\n",
       "      <td>56nd7</td>\n",
       "      <td>confident</td>\n",
       "      <td>AI</td>\n",
       "      <td>confident</td>\n",
       "      <td>56nd7</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25725</th>\n",
       "      <td>565</td>\n",
       "      <td>4</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>lmxbd</td>\n",
       "      <td>lmxbd</td>\n",
       "      <td>confident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>neutral</td>\n",
       "      <td>lmxbd</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25737</th>\n",
       "      <td>565</td>\n",
       "      <td>5</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>ppfuk</td>\n",
       "      <td>ppfuk</td>\n",
       "      <td>confident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>confident</td>\n",
       "      <td>ppfuk</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25749</th>\n",
       "      <td>565</td>\n",
       "      <td>6</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>sybe3</td>\n",
       "      <td>sybe3</td>\n",
       "      <td>confident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>confident</td>\n",
       "      <td>sybe3</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25754</th>\n",
       "      <td>565</td>\n",
       "      <td>7</td>\n",
       "      <td>captcha-images</td>\n",
       "      <td>efcns</td>\n",
       "      <td>f3cx2</td>\n",
       "      <td>unconfident</td>\n",
       "      <td>AI</td>\n",
       "      <td>neutral</td>\n",
       "      <td>f3cx2</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25763</th>\n",
       "      <td>565</td>\n",
       "      <td>8</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>c9mxn</td>\n",
       "      <td>c3mxn</td>\n",
       "      <td>confident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>confident</td>\n",
       "      <td>c9mxn</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25773</th>\n",
       "      <td>565</td>\n",
       "      <td>9</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>4fjud</td>\n",
       "      <td>4fjud</td>\n",
       "      <td>confident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>confident</td>\n",
       "      <td>4fjud</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25783</th>\n",
       "      <td>565</td>\n",
       "      <td>10</td>\n",
       "      <td>captcha-09az</td>\n",
       "      <td>yg5bb</td>\n",
       "      <td>mg5bb</td>\n",
       "      <td>unconfident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>neutral</td>\n",
       "      <td>yg5bb</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25797</th>\n",
       "      <td>565</td>\n",
       "      <td>11</td>\n",
       "      <td>captcha-images</td>\n",
       "      <td>twzbi</td>\n",
       "      <td>x8cxc</td>\n",
       "      <td>unconfident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>neutral</td>\n",
       "      <td>tw2bi</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25809</th>\n",
       "      <td>565</td>\n",
       "      <td>12</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>ypwrq</td>\n",
       "      <td>ypwrq</td>\n",
       "      <td>confident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>confident</td>\n",
       "      <td>ypwrq</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25813</th>\n",
       "      <td>565</td>\n",
       "      <td>13</td>\n",
       "      <td>captcha-09az</td>\n",
       "      <td>4ynf3</td>\n",
       "      <td>4ynf3</td>\n",
       "      <td>confident</td>\n",
       "      <td>AI</td>\n",
       "      <td>neutral</td>\n",
       "      <td>4ynf3</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25816</th>\n",
       "      <td>565</td>\n",
       "      <td>14</td>\n",
       "      <td>captcha-09az</td>\n",
       "      <td>5nnff</td>\n",
       "      <td>5mnff</td>\n",
       "      <td>unconfident</td>\n",
       "      <td>AI</td>\n",
       "      <td>neutral</td>\n",
       "      <td>5mnff</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25821</th>\n",
       "      <td>565</td>\n",
       "      <td>15</td>\n",
       "      <td>captcha-09az</td>\n",
       "      <td>mcg43</td>\n",
       "      <td>mcg43</td>\n",
       "      <td>confident</td>\n",
       "      <td>AI</td>\n",
       "      <td>neutral</td>\n",
       "      <td>mcg43</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25830</th>\n",
       "      <td>565</td>\n",
       "      <td>16</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>8r3tb</td>\n",
       "      <td>8r3tb</td>\n",
       "      <td>confident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>neutral</td>\n",
       "      <td>8r3tb</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25844</th>\n",
       "      <td>565</td>\n",
       "      <td>17</td>\n",
       "      <td>captcha-09az</td>\n",
       "      <td>d7c5x</td>\n",
       "      <td>d7c5x</td>\n",
       "      <td>confident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>confident</td>\n",
       "      <td>d7c5x</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25851</th>\n",
       "      <td>565</td>\n",
       "      <td>18</td>\n",
       "      <td>capital-color</td>\n",
       "      <td>ykvv2</td>\n",
       "      <td>ykvv2</td>\n",
       "      <td>confident</td>\n",
       "      <td>AI</td>\n",
       "      <td>confident</td>\n",
       "      <td>ykvv2</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25858</th>\n",
       "      <td>565</td>\n",
       "      <td>19</td>\n",
       "      <td>captcha-images</td>\n",
       "      <td>urjaz</td>\n",
       "      <td>nnccc</td>\n",
       "      <td>unconfident</td>\n",
       "      <td>SELF</td>\n",
       "      <td>neutral</td>\n",
       "      <td>urjaz</td>\n",
       "      <td>SelectiveCueUser3</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       user_id  episode_id    dataset_name ground_truth ai_answer  \\\n",
       "id                                                                  \n",
       "25655      565           0    captcha-09az        fg38b     fg38b   \n",
       "25677      565           1   capital-color        nrby8     nrby8   \n",
       "25706      565           2   capital-color        s3ych     s3ych   \n",
       "25714      565           3    captcha-09az        5bnd7     56nd7   \n",
       "25725      565           4   capital-color        lmxbd     lmxbd   \n",
       "25737      565           5   capital-color        ppfuk     ppfuk   \n",
       "25749      565           6   capital-color        sybe3     sybe3   \n",
       "25754      565           7  captcha-images        efcns     f3cx2   \n",
       "25763      565           8   capital-color        c9mxn     c3mxn   \n",
       "25773      565           9   capital-color        4fjud     4fjud   \n",
       "25783      565          10    captcha-09az        yg5bb     mg5bb   \n",
       "25797      565          11  captcha-images        twzbi     x8cxc   \n",
       "25809      565          12   capital-color        ypwrq     ypwrq   \n",
       "25813      565          13    captcha-09az        4ynf3     4ynf3   \n",
       "25816      565          14    captcha-09az        5nnff     5mnff   \n",
       "25821      565          15    captcha-09az        mcg43     mcg43   \n",
       "25830      565          16   capital-color        8r3tb     8r3tb   \n",
       "25844      565          17    captcha-09az        d7c5x     d7c5x   \n",
       "25851      565          18   capital-color        ykvv2     ykvv2   \n",
       "25858      565          19  captcha-images        urjaz     nnccc   \n",
       "\n",
       "      potential_cue user_decision system_action final_answer  \\\n",
       "id                                                             \n",
       "25655     confident            AI       neutral        fg38b   \n",
       "25677     confident          SELF       neutral        nrby8   \n",
       "25706   unconfident          SELF   unconfident        s3ych   \n",
       "25714     confident            AI     confident        56nd7   \n",
       "25725     confident          SELF       neutral        lmxbd   \n",
       "25737     confident          SELF     confident        ppfuk   \n",
       "25749     confident          SELF     confident        sybe3   \n",
       "25754   unconfident            AI       neutral        f3cx2   \n",
       "25763     confident          SELF     confident        c9mxn   \n",
       "25773     confident          SELF     confident        4fjud   \n",
       "25783   unconfident          SELF       neutral        yg5bb   \n",
       "25797   unconfident          SELF       neutral        tw2bi   \n",
       "25809     confident          SELF     confident        ypwrq   \n",
       "25813     confident            AI       neutral        4ynf3   \n",
       "25816   unconfident            AI       neutral        5mnff   \n",
       "25821     confident            AI       neutral        mcg43   \n",
       "25830     confident          SELF       neutral        8r3tb   \n",
       "25844     confident          SELF     confident        d7c5x   \n",
       "25851     confident            AI     confident        ykvv2   \n",
       "25858   unconfident          SELF       neutral        urjaz   \n",
       "\n",
       "              user_class  ai_correct  final_answer_correct  cue_shown  \\\n",
       "id                                                                      \n",
       "25655  SelectiveCueUser3        True                  True      False   \n",
       "25677  SelectiveCueUser3        True                  True      False   \n",
       "25706  SelectiveCueUser3        True                  True       True   \n",
       "25714  SelectiveCueUser3       False                 False       True   \n",
       "25725  SelectiveCueUser3        True                  True      False   \n",
       "25737  SelectiveCueUser3        True                  True       True   \n",
       "25749  SelectiveCueUser3        True                  True       True   \n",
       "25754  SelectiveCueUser3       False                 False      False   \n",
       "25763  SelectiveCueUser3       False                  True       True   \n",
       "25773  SelectiveCueUser3        True                  True       True   \n",
       "25783  SelectiveCueUser3       False                  True      False   \n",
       "25797  SelectiveCueUser3       False                 False      False   \n",
       "25809  SelectiveCueUser3        True                  True       True   \n",
       "25813  SelectiveCueUser3        True                  True      False   \n",
       "25816  SelectiveCueUser3       False                 False      False   \n",
       "25821  SelectiveCueUser3        True                  True      False   \n",
       "25830  SelectiveCueUser3        True                  True      False   \n",
       "25844  SelectiveCueUser3        True                  True       True   \n",
       "25851  SelectiveCueUser3        True                  True       True   \n",
       "25858  SelectiveCueUser3       False                  True      False   \n",
       "\n",
       "       assignment_correct  \n",
       "id                         \n",
       "25655                True  \n",
       "25677               False  \n",
       "25706               False  \n",
       "25714               False  \n",
       "25725               False  \n",
       "25737               False  \n",
       "25749               False  \n",
       "25754               False  \n",
       "25763                True  \n",
       "25773               False  \n",
       "25783                True  \n",
       "25797                True  \n",
       "25809               False  \n",
       "25813                True  \n",
       "25816               False  \n",
       "25821                True  \n",
       "25830               False  \n",
       "25844               False  \n",
       "25851                True  \n",
       "25858                True  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Figure 8 (b) (unsuccessful example)\n",
    "display(_target.query(\"episode_id < 20\").sort_values('episode_id').query('user_id == 565')[columns])"
   ]
  }
 ],
 "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": "3.8.10"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}