{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The file with the updated p_dup and p_map columns has been saved: 0.class_document/knn_tfidf/1/test_p_c_r.csv\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "group_number = 1\n",
    "method_name='knn_tfidf'\n",
    "# Read the test file\n",
    "test_path = f'0.class_document/{method_name}/{group_number}/test_p_c_r.csv'\n",
    "df = pd.read_csv(test_path)\n",
    "\n",
    "# Concatenate p_thing and p_property into p_tp in the test data\n",
    "df['p_tp'] = df['p_thing'] + \" \" + df['p_property']\n",
    "\n",
    "# Read the train_all file\n",
    "train_all_path = f'../../data_preprocess/dataset/{group_number}/train_all.csv'\n",
    "train_all_df = pd.read_csv(train_all_path)\n",
    "\n",
    "# Concatenate thing and property into tp in the train_all data\n",
    "train_all_df['tp'] = train_all_df['thing'] + \" \" + train_all_df['property']\n",
    "\n",
    "# Initialize the p_map column in the test data\n",
    "df['p_map'] = 0\n",
    "\n",
    "# Group by ships_idx and then group p_tp within each ships_idx group\n",
    "grouped = df.groupby('ships_idx')['p_tp']\n",
    "\n",
    "# Iterate through each ships_idx group\n",
    "for ships_idx, group in grouped:\n",
    "    # Count the occurrences of each p_tp within the test group\n",
    "    p_tp_counts = group.value_counts()\n",
    "    \n",
    "    # Assign the count as an integer to p_dup for rows with the corresponding p_tp within the group\n",
    "    for p_tp, count in p_tp_counts.items():\n",
    "        # Update p_dup\n",
    "        df.loc[(df['ships_idx'] == ships_idx) & (df['p_tp'] == p_tp), 'p_dup'] = int(count)\n",
    "        \n",
    "        # Calculate p_map by counting matching tp in train_all_df\n",
    "        p_map_count = train_all_df['tp'].eq(p_tp).sum()\n",
    "        df.loc[(df['ships_idx'] == ships_idx) & (df['p_tp'] == p_tp), 'p_map'] = int(p_map_count)\n",
    "\n",
    "# Save the modified DataFrame\n",
    "output_path = f'0.class_document/{method_name}/{group_number}/test_p_c_r.csv'\n",
    "df.to_csv(output_path, index=False, encoding='utf-8-sig')\n",
    "\n",
    "print(\"The file with the updated p_dup and p_map columns has been saved:\", output_path)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "torch",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}