File size: 5,857 Bytes
0758d29
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2b1f5378",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import pandas as pd\n",
    "import itertools\n",
    "import os\n",
    "\n",
    "# Load data files\n",
    "annotator_df = pd.read_csv(\"./trac4_PER_train.csv\")\n",
    "df = pd.read_csv(\"./trac3_EMP_train.csv\", on_bad_lines='skip')\n",
    "article_df = pd.read_csv(\"./articles_adobe_AMT.csv\")\n",
    "\n",
    "# Mappings for structured user profile\n",
    "gender_map = {1: 'Male', 2: 'Female', 5: 'Other'}\n",
    "race_map = {1: 'White', 2: 'Hispanic / Latino', 3: 'Black / African American', 4: 'Native American / American Indian', 5: 'Asian / Pacific Islander', 6: 'Other'}\n",
    "education_map = {\n",
    "    1: 'a diploma less than a high school',\n",
    "    2: 'High school degree or diploma',\n",
    "    3: 'went to Technical / Vocational School',\n",
    "    4: 'went to college but did not get a degree',\n",
    "    5: 'Two year associate degree',\n",
    "    6: 'College or university degree',\n",
    "    7: 'Postgraduate / professional degree'\n",
    "}\n",
    "\n",
    "# Build annotator dicts\n",
    "annotator_text_dict = {}\n",
    "annotator_structured_dict = {}\n",
    "for idx, row in annotator_df.iterrows():\n",
    "    persona = []\n",
    "    structured = {}\n",
    "    gender = gender_map.get(row['gender'], 'Unknown')\n",
    "    race = race_map.get(row['race'], 'Unknown')\n",
    "    age = row['age']\n",
    "    education = education_map.get(row['education'], 'Unknown')\n",
    "    income = row['income']\n",
    "    personality = {\n",
    "        \"openness\": row['personality_openess'],\n",
    "        \"conscientiousness\": row['personality_conscientiousness'],\n",
    "        \"extraversion\": row['personality_extraversion'],\n",
    "        \"agreeableness\": row['personality_agreeableness'],\n",
    "        \"stability\": row['personality_stability']\n",
    "    }\n",
    "    persona.append(f\"The person is {gender.lower()}.\")\n",
    "    persona.append(f\"Racially, the person is {race.lower()}.\")\n",
    "    persona.append(f\"The person is {age} years old.\")\n",
    "    persona.append(f\"The person has a {education.lower()}.\")\n",
    "    persona.append(f\"The person earns {income} dollar per year.\")\n",
    "    persona.append(\n",
    "        f\"According to the Big Five personality test, on a scale of 10, the person has scored {personality['openness']} in openness, {personality['conscientiousness']} in conscientiousness, {personality['extraversion']} in extraversion, {personality['agreeableness']} in agreeableness, and {personality['stability']} in stability.\"\n",
    "    )\n",
    "    persona = [p for p in persona if \"nan\" not in p]\n",
    "    persona_text = \" \".join(persona)\n",
    "    annotator_text_dict[row['person_id']] = persona_text\n",
    "    structured = {\n",
    "        \"gender\": gender,\n",
    "        \"race\": race,\n",
    "        \"age\": age,\n",
    "        \"education\": education,\n",
    "        \"income\": income,\n",
    "        \"personality\": personality\n",
    "    }\n",
    "    annotator_structured_dict[row['person_id']] = structured\n",
    "\n",
    "# Prepare dataset\n",
    "dataset = []\n",
    "for idx, row in df.iterrows():\n",
    "    person_id = row['person_id']\n",
    "    article_id = row['article_id']\n",
    "    # Get user profile\n",
    "    user_profile_text = annotator_text_dict.get(person_id, \"\")\n",
    "    user_profile_structured = annotator_structured_dict.get(person_id, {})\n",
    "    # Get article\n",
    "    article_row = article_df[article_df['article_id'] == article_id]\n",
    "    if article_row.empty:\n",
    "        continue\n",
    "    article_text = article_row['text'].values[0]\n",
    "    # Input prompt (as in ec.py, with_persona)\n",
    "    prompt = (\n",
    "        f\"{article_text}\"\n",
    "    )\n",
    "    # Output: the user's essay/response\n",
    "    output = row['person_essay']\n",
    "    # Compose item\n",
    "    item = {\n",
    "        \"user_id\": f\"ec_{person_id}\",\n",
    "        \"profile_text\": user_profile_text,\n",
    "        \"profile\": user_profile_structured,\n",
    "        \"article\": article_text,\n",
    "        # \"input\": prompt,\n",
    "        \"essay\": output\n",
    "    }\n",
    "    dataset.append(item)\n",
    "\n",
    "# Save to JSON\n",
    "# with open(\"ec_dataset.json\", \"w\", encoding=\"utf-8\") as f:\n",
    "#     json.dump(dataset, f, ensure_ascii=False, indent=2)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "cb69aa41",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "974"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "13f8cca9",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('/home/zxtan/text-to-lora/data_p13n/EC/ec_dataset.jsonl', 'w') as f:\n",
    "    for line in dataset:\n",
    "        f.write(json.dumps(line) + '\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "46944ebd",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pytorch",
   "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": "3.11.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}