{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import gc\n",
    "sns.set_style(\"darkgrid\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "fpmms = pd.read_parquet('../data/fpmms.parquet')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>id</th>\n",
       "      <th>currentAnswer</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0x0020d13c89140b47e10db54cbd53852b90bc1391</td>\n",
       "      <td>No</td>\n",
       "      <td>Will the Francis Scott Key Bridge in Baltimore...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0x003ae5e007cc38b3f86b0ed7c82f938a1285ac07</td>\n",
       "      <td>No</td>\n",
       "      <td>Will FC Saarbrucken reach the final of the Ger...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0x004c8d4c619dc6b9caa940f5ea7ef699ae85359c</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Will the pro-life activists convicted for 'con...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0x005e3f7a90585acbec807425a750fbba1d0c2b5c</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Will Apple announce the release of a new M4 ch...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0x0094fa304017d5c2b355790e2976f769ea600492</td>\n",
       "      <td>No</td>\n",
       "      <td>Will the Hisense U8K be considered a top-tier ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           id currentAnswer  \\\n",
       "0  0x0020d13c89140b47e10db54cbd53852b90bc1391            No   \n",
       "1  0x003ae5e007cc38b3f86b0ed7c82f938a1285ac07            No   \n",
       "2  0x004c8d4c619dc6b9caa940f5ea7ef699ae85359c           Yes   \n",
       "3  0x005e3f7a90585acbec807425a750fbba1d0c2b5c           Yes   \n",
       "4  0x0094fa304017d5c2b355790e2976f769ea600492            No   \n",
       "\n",
       "                                               title  \n",
       "0  Will the Francis Scott Key Bridge in Baltimore...  \n",
       "1  Will FC Saarbrucken reach the final of the Ger...  \n",
       "2  Will the pro-life activists convicted for 'con...  \n",
       "3  Will Apple announce the release of a new M4 ch...  \n",
       "4  Will the Hisense U8K be considered a top-tier ...  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fpmms.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "currentAnswer\n",
       "No       2170\n",
       "Yes      1500\n",
       "no          1\n",
       "False       1\n",
       "IND         1\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fpmms.currentAnswer.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 3673 entries, 0 to 3672\n",
      "Data columns (total 3 columns):\n",
      " #   Column         Non-Null Count  Dtype \n",
      "---  ------         --------------  ----- \n",
      " 0   id             3673 non-null   object\n",
      " 1   currentAnswer  3673 non-null   object\n",
      " 2   title          3673 non-null   object\n",
      "dtypes: object(3)\n",
      "memory usage: 86.2+ KB\n"
     ]
    }
   ],
   "source": [
    "fpmms.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_trades = pd.read_parquet('../data/all_trades_profitability.parquet')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "is_invalid\n",
       "False    23830\n",
       "True      3877\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_trades.is_invalid.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "24722"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mask = (all_trades[\"is_invalid\"] & all_trades[\"redeemed\"])\n",
    "filtered_trades = all_trades[~mask]\n",
    "len(filtered_trades)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "27707"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(all_trades)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "24722"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(filtered_trades)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "winning_trade\n",
       "True     13133\n",
       "False    11589\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_trades.winning_trade.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/gp/02mb1d514ng739czlxw1lhh00000gn/T/ipykernel_82376/982645160.py:1: 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",
      "  filtered_trades[\"creation_timestamp\"] = pd.to_datetime(filtered_trades[\"creation_timestamp\"])\n"
     ]
    }
   ],
   "source": [
    "filtered_trades[\"creation_timestamp\"] = pd.to_datetime(filtered_trades[\"creation_timestamp\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "current_answer\n",
       " 1    13016\n",
       " 0    10814\n",
       "-1      892\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_trades.current_answer.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "203"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(list(all_trades.trader_address.unique()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "27707"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(all_trades)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/gp/02mb1d514ng739czlxw1lhh00000gn/T/ipykernel_70112/183699308.py:1: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.\n",
      "  all_trades['month_year'] = all_trades['creation_timestamp'].dt.to_period('M').astype(str)\n",
      "/var/folders/gp/02mb1d514ng739czlxw1lhh00000gn/T/ipykernel_70112/183699308.py:2: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.\n",
      "  all_trades['month_year_week'] = all_trades['creation_timestamp'].dt.to_period('W').astype(str)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "winning_trade\n",
       "0    14574\n",
       "1    13133\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_trades['month_year'] = all_trades['creation_timestamp'].dt.to_period('M').astype(str)\n",
    "all_trades['month_year_week'] = all_trades['creation_timestamp'].dt.to_period('W').astype(str)\n",
    "all_trades['winning_trade'] = all_trades['winning_trade'].astype(int)\n",
    "all_trades.winning_trade.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>month_year_week</th>\n",
       "      <th>winning_trade</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2024-04-22/2024-04-28</td>\n",
       "      <td>60.465116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2024-04-29/2024-05-05</td>\n",
       "      <td>53.887043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024-05-06/2024-05-12</td>\n",
       "      <td>49.626201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2024-05-13/2024-05-19</td>\n",
       "      <td>47.931617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2024-05-20/2024-05-26</td>\n",
       "      <td>46.209810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2024-05-27/2024-06-02</td>\n",
       "      <td>41.855369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2024-06-03/2024-06-09</td>\n",
       "      <td>43.714888</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2024-06-10/2024-06-16</td>\n",
       "      <td>46.697039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2024-06-17/2024-06-23</td>\n",
       "      <td>52.762120</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         month_year_week  winning_trade\n",
       "0  2024-04-22/2024-04-28      60.465116\n",
       "1  2024-04-29/2024-05-05      53.887043\n",
       "2  2024-05-06/2024-05-12      49.626201\n",
       "3  2024-05-13/2024-05-19      47.931617\n",
       "4  2024-05-20/2024-05-26      46.209810\n",
       "5  2024-05-27/2024-06-02      41.855369\n",
       "6  2024-06-03/2024-06-09      43.714888\n",
       "7  2024-06-10/2024-06-16      46.697039\n",
       "8  2024-06-17/2024-06-23      52.762120"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "winning_trades = all_trades.groupby(['month_year_week'])['winning_trade'].sum() / all_trades.groupby(['month_year_week'])['winning_trade'].count() * 100\n",
    "# winning_trades is a series, give it a dataframe\n",
    "winning_trades = winning_trades.reset_index()\n",
    "winning_trades.columns = winning_trades.columns.astype(str)\n",
    "winning_trades.columns = ['month_year_week', 'winning_trade']\n",
    "winning_trades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>month_year_week</th>\n",
       "      <th>winning_trade</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2024-06-03/2024-06-09</td>\n",
       "      <td>43.714888</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         month_year_week  winning_trade\n",
       "6  2024-06-03/2024-06-09      43.714888"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "that_week = winning_trades[winning_trades[\"month_year_week\"]==\"2024-06-03/2024-06-09\"]\n",
    "that_week"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>month_year_week</th>\n",
       "      <th>sum</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2024-04-22/2024-04-28</td>\n",
       "      <td>26</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2024-04-29/2024-05-05</td>\n",
       "      <td>1622</td>\n",
       "      <td>3010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024-05-06/2024-05-12</td>\n",
       "      <td>2788</td>\n",
       "      <td>5618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2024-05-13/2024-05-19</td>\n",
       "      <td>2271</td>\n",
       "      <td>4738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2024-05-20/2024-05-26</td>\n",
       "      <td>1969</td>\n",
       "      <td>4261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2024-05-27/2024-06-02</td>\n",
       "      <td>1719</td>\n",
       "      <td>4107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2024-06-03/2024-06-09</td>\n",
       "      <td>1245</td>\n",
       "      <td>2848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2024-06-10/2024-06-16</td>\n",
       "      <td>1025</td>\n",
       "      <td>2195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2024-06-17/2024-06-23</td>\n",
       "      <td>468</td>\n",
       "      <td>887</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         month_year_week   sum  count\n",
       "0  2024-04-22/2024-04-28    26     43\n",
       "1  2024-04-29/2024-05-05  1622   3010\n",
       "2  2024-05-06/2024-05-12  2788   5618\n",
       "3  2024-05-13/2024-05-19  2271   4738\n",
       "4  2024-05-20/2024-05-26  1969   4261\n",
       "5  2024-05-27/2024-06-02  1719   4107\n",
       "6  2024-06-03/2024-06-09  1245   2848\n",
       "7  2024-06-10/2024-06-16  1025   2195\n",
       "8  2024-06-17/2024-06-23   468    887"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "winning_trades2 = all_trades.groupby(['month_year_week'])['winning_trade'].agg([\"sum\",\"count\"]).reset_index()\n",
    "winning_trades2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>month_year_week</th>\n",
       "      <th>sum</th>\n",
       "      <th>count</th>\n",
       "      <th>winning_trade</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2024-06-03/2024-06-09</td>\n",
       "      <td>1245</td>\n",
       "      <td>2848</td>\n",
       "      <td>43.714888</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         month_year_week   sum  count  winning_trade\n",
       "6  2024-06-03/2024-06-09  1245   2848      43.714888"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "that_week = winning_trades2[winning_trades2[\"month_year_week\"]==\"2024-06-03/2024-06-09\"]\n",
    "that_week"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "INC_TOOLS = [\n",
    "    \"prediction-online\",\n",
    "    \"prediction-offline\",\n",
    "    \"claude-prediction-online\",\n",
    "    \"claude-prediction-offline\",\n",
    "    \"prediction-offline-sme\",\n",
    "    \"prediction-online-sme\",\n",
    "    \"prediction-request-rag\",\n",
    "    \"prediction-request-reasoning\",\n",
    "    \"prediction-url-cot-claude\",\n",
    "    \"prediction-request-rag-claude\",\n",
    "    \"prediction-request-reasoning-claude\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "tools = pd.read_parquet('../data/tools.parquet')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 127674 entries, 0 to 127673\n",
      "Data columns (total 22 columns):\n",
      " #   Column                   Non-Null Count   Dtype  \n",
      "---  ------                   --------------   -----  \n",
      " 0   request_id               127674 non-null  object \n",
      " 1   request_block            127674 non-null  int64  \n",
      " 2   prompt_request           127674 non-null  object \n",
      " 3   tool                     127674 non-null  object \n",
      " 4   nonce                    127674 non-null  object \n",
      " 5   trader_address           127674 non-null  object \n",
      " 6   deliver_block            127674 non-null  int64  \n",
      " 7   error                    127668 non-null  float64\n",
      " 8   error_message            19534 non-null   object \n",
      " 9   prompt_response          120607 non-null  object \n",
      " 10  mech_address             127674 non-null  object \n",
      " 11  p_yes                    108134 non-null  float64\n",
      " 12  p_no                     108134 non-null  float64\n",
      " 13  confidence               108134 non-null  float64\n",
      " 14  info_utility             108134 non-null  float64\n",
      " 15  vote                     94137 non-null   object \n",
      " 16  win_probability          108134 non-null  float64\n",
      " 17  title                    118074 non-null  object \n",
      " 18  currentAnswer            88330 non-null   object \n",
      " 19  request_time             127674 non-null  object \n",
      " 20  request_month_year       127674 non-null  object \n",
      " 21  request_month_year_week  127674 non-null  object \n",
      "dtypes: float64(6), int64(2), object(14)\n",
      "memory usage: 21.4+ MB\n"
     ]
    }
   ],
   "source": [
    "tools.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "currentAnswer\n",
       "No     51140\n",
       "Yes    37190\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tools.currentAnswer.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "127674"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(tools)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "tools_inc = tools[tools['tool'].isin(INC_TOOLS)]\n",
    "tools_non_error = tools_inc[tools_inc['error'] != 1]\n",
    "tools_non_error.loc[:, 'currentAnswer'] = tools_non_error['currentAnswer'].replace({'no': 'No', 'yes': 'Yes'})\n",
    "tools_non_error = tools_non_error[tools_non_error['currentAnswer'].isin(['Yes', 'No'])]\n",
    "tools_non_error = tools_non_error[tools_non_error['vote'].isin(['Yes', 'No'])]\n",
    "tools_non_error['win'] = (tools_non_error['currentAnswer'] == tools_non_error['vote']).astype(int)\n",
    "tools_non_error.columns = tools_non_error.columns.astype(str)\n",
    "wins = tools_non_error.groupby(['tool', 'request_month_year_week', 'win']).size().unstack().fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "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>win</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tool</th>\n",
       "      <th>request_month_year_week</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">claude-prediction-offline</th>\n",
       "      <th>2024-04-22/2024-04-28</th>\n",
       "      <td>14.0</td>\n",
       "      <td>23.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-04-29/2024-05-05</th>\n",
       "      <td>34.0</td>\n",
       "      <td>99.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-05-06/2024-05-12</th>\n",
       "      <td>22.0</td>\n",
       "      <td>34.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-05-13/2024-05-19</th>\n",
       "      <td>40.0</td>\n",
       "      <td>52.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-05-20/2024-05-26</th>\n",
       "      <td>18.0</td>\n",
       "      <td>52.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">prediction-url-cot-claude</th>\n",
       "      <th>2024-05-06/2024-05-12</th>\n",
       "      <td>67.0</td>\n",
       "      <td>91.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-05-13/2024-05-19</th>\n",
       "      <td>28.0</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-05-20/2024-05-26</th>\n",
       "      <td>64.0</td>\n",
       "      <td>145.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-05-27/2024-06-02</th>\n",
       "      <td>81.0</td>\n",
       "      <td>112.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-06-03/2024-06-09</th>\n",
       "      <td>7.0</td>\n",
       "      <td>41.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>91 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "win                                                   0      1\n",
       "tool                      request_month_year_week             \n",
       "claude-prediction-offline 2024-04-22/2024-04-28    14.0   23.0\n",
       "                          2024-04-29/2024-05-05    34.0   99.0\n",
       "                          2024-05-06/2024-05-12    22.0   34.0\n",
       "                          2024-05-13/2024-05-19    40.0   52.0\n",
       "                          2024-05-20/2024-05-26    18.0   52.0\n",
       "...                                                 ...    ...\n",
       "prediction-url-cot-claude 2024-05-06/2024-05-12    67.0   91.0\n",
       "                          2024-05-13/2024-05-19    28.0   43.0\n",
       "                          2024-05-20/2024-05-26    64.0  145.0\n",
       "                          2024-05-27/2024-06-02    81.0  112.0\n",
       "                          2024-06-03/2024-06-09     7.0   41.0\n",
       "\n",
       "[91 rows x 2 columns]"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wins"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "186"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selected_traders = list(tools.trader_address.unique())\n",
    "len(selected_traders)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "182"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(list(tools_non_error.trader_address.unique()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10817"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(tools)-len(tools_inc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11778"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tools_week = tools_non_error[tools_non_error[\"request_month_year_week\"]==\"2024-06-03/2024-06-09\"]\n",
    "len(tools_week)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_trades = all_trades.loc[all_trades[\"trader_address\"].isin(selected_traders)]\n",
    "len(filtered_trades)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_addresses = list(all_trades.trader_address.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "for a in all_addresses:\n",
    "    if a in selected_traders:\n",
    "        print(\"found\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "for a in selected_traders:\n",
    "    if a in all_addresses:\n",
    "        print(\"found\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_tools = tools[tools[\"trader_address\"].isin(all_addresses)]\n",
    "len(filtered_tools)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    27707.000000\n",
       "mean         3.912224\n",
       "std          4.622220\n",
       "min          0.000000\n",
       "25%          1.000000\n",
       "50%          2.000000\n",
       "75%          5.000000\n",
       "max         66.000000\n",
       "Name: num_mech_calls, dtype: float64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_trades.num_mech_calls.describe()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "market_creator",
   "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": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}