File size: 14,835 Bytes
fd59c75 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 |
{
"cells": [
{
"cell_type": "code",
"execution_count": 43,
"metadata": {},
"outputs": [],
"source": [
"#dependencies:\n",
"import pandas as pd\n",
"\n",
"import torch\n",
"from transformers import GPT2Tokenizer\n",
"\n",
"from trl import AutoModelForCausalLMWithValueHead, PPOConfig, PPOTrainer"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b8a22b8d60c0417eafbf554832398287",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Resolving data files: 0%| | 0/18 [00:00<?, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b83d2624c2b14986a8297821460225ab",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Resolving data files: 0%| | 0/18 [00:00<?, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b4304c0f48cb472589b5e80d3a42cba2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Resolving data files: 0%| | 0/18 [00:00<?, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"#loading datasets:\n",
"from datasets import load_dataset\n",
"\n",
"ds = load_dataset(\"stanfordnlp/SHP\", split='train')"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Index(['post_id', 'domain', 'upvote_ratio', 'history', 'c_root_id_A',\n",
" 'c_root_id_B', 'created_at_utc_A', 'created_at_utc_B', 'score_A',\n",
" 'score_B', 'human_ref_A', 'human_ref_B', 'labels', 'seconds_difference',\n",
" 'score_ratio'],\n",
" dtype='object')\n"
]
}
],
"source": [
"df = ds.to_pandas()\n",
"print(df.columns)\n"
]
},
{
"cell_type": "code",
"execution_count": 46,
"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>upvote_ratio</th>\n",
" <th>history</th>\n",
" <th>score_A</th>\n",
" <th>score_B</th>\n",
" <th>human_ref_A</th>\n",
" <th>human_ref_B</th>\n",
" <th>labels</th>\n",
" <th>score_ratio</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>0.99</td>\n",
" <td>In an interview right before receiving the 201...</td>\n",
" <td>52</td>\n",
" <td>54</td>\n",
" <td>Currently wrapping up my PhD. There is a stark...</td>\n",
" <td>It’s ironic to me that research has shown that...</td>\n",
" <td>0</td>\n",
" <td>1.038462</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>0.95</td>\n",
" <td>If any professor is reading this: please do no...</td>\n",
" <td>5</td>\n",
" <td>17</td>\n",
" <td>And when your teacher doesn't listen or pay at...</td>\n",
" <td>I'm pretty strict on time, to the point where ...</td>\n",
" <td>0</td>\n",
" <td>3.400000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>0.95</td>\n",
" <td>If any professor is reading this: please do no...</td>\n",
" <td>5</td>\n",
" <td>7</td>\n",
" <td>Profs can be oblivious? What’s new!</td>\n",
" <td>This sounds like a problem with a specific pro...</td>\n",
" <td>0</td>\n",
" <td>1.400000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>0.95</td>\n",
" <td>If any professor is reading this: please do no...</td>\n",
" <td>7</td>\n",
" <td>5</td>\n",
" <td>This sounds like a problem with a specific pro...</td>\n",
" <td>And when your teacher doesn't listen or pay at...</td>\n",
" <td>1</td>\n",
" <td>1.400000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>0.95</td>\n",
" <td>If any professor is reading this: please do no...</td>\n",
" <td>6</td>\n",
" <td>7</td>\n",
" <td>This would be totally unacceptable in my class...</td>\n",
" <td>This sounds like a problem with a specific pro...</td>\n",
" <td>0</td>\n",
" <td>1.166667</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>348713</th>\n",
" <td>0.94</td>\n",
" <td>Can I get in trouble for giving my neighbor hi...</td>\n",
" <td>7</td>\n",
" <td>25</td>\n",
" <td>Just put up a fence. Legally he isn't responsi...</td>\n",
" <td>Whatever you do, don't cut his trees down.</td>\n",
" <td>0</td>\n",
" <td>3.571429</td>\n",
" </tr>\n",
" <tr>\n",
" <th>348714</th>\n",
" <td>0.94</td>\n",
" <td>Can I get in trouble for giving my neighbor hi...</td>\n",
" <td>2</td>\n",
" <td>25</td>\n",
" <td>If OP pays someone to clean his yard, and then...</td>\n",
" <td>Whatever you do, don't cut his trees down.</td>\n",
" <td>0</td>\n",
" <td>12.500000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>348715</th>\n",
" <td>0.94</td>\n",
" <td>Can I get in trouble for giving my neighbor hi...</td>\n",
" <td>9</td>\n",
" <td>7</td>\n",
" <td>My observation is that both of you are idiots...</td>\n",
" <td>Are you Rand Paul's neighbor? https://www.gq....</td>\n",
" <td>1</td>\n",
" <td>1.285714</td>\n",
" </tr>\n",
" <tr>\n",
" <th>348716</th>\n",
" <td>0.94</td>\n",
" <td>Can I get in trouble for giving my neighbor hi...</td>\n",
" <td>9</td>\n",
" <td>7</td>\n",
" <td>My observation is that both of you are idiots...</td>\n",
" <td>Just put up a fence. Legally he isn't responsi...</td>\n",
" <td>1</td>\n",
" <td>1.285714</td>\n",
" </tr>\n",
" <tr>\n",
" <th>348717</th>\n",
" <td>0.94</td>\n",
" <td>Can I get in trouble for giving my neighbor hi...</td>\n",
" <td>7</td>\n",
" <td>2</td>\n",
" <td>Capture his acts on camera. Collect and bag l...</td>\n",
" <td>If OP pays someone to clean his yard, and then...</td>\n",
" <td>1</td>\n",
" <td>3.500000</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>348718 rows × 8 columns</p>\n",
"</div>"
],
"text/plain": [
" upvote_ratio history \\\n",
"0 0.99 In an interview right before receiving the 201... \n",
"1 0.95 If any professor is reading this: please do no... \n",
"2 0.95 If any professor is reading this: please do no... \n",
"3 0.95 If any professor is reading this: please do no... \n",
"4 0.95 If any professor is reading this: please do no... \n",
"... ... ... \n",
"348713 0.94 Can I get in trouble for giving my neighbor hi... \n",
"348714 0.94 Can I get in trouble for giving my neighbor hi... \n",
"348715 0.94 Can I get in trouble for giving my neighbor hi... \n",
"348716 0.94 Can I get in trouble for giving my neighbor hi... \n",
"348717 0.94 Can I get in trouble for giving my neighbor hi... \n",
"\n",
" score_A score_B human_ref_A \\\n",
"0 52 54 Currently wrapping up my PhD. There is a stark... \n",
"1 5 17 And when your teacher doesn't listen or pay at... \n",
"2 5 7 Profs can be oblivious? What’s new! \n",
"3 7 5 This sounds like a problem with a specific pro... \n",
"4 6 7 This would be totally unacceptable in my class... \n",
"... ... ... ... \n",
"348713 7 25 Just put up a fence. Legally he isn't responsi... \n",
"348714 2 25 If OP pays someone to clean his yard, and then... \n",
"348715 9 7 My observation is that both of you are idiots... \n",
"348716 9 7 My observation is that both of you are idiots... \n",
"348717 7 2 Capture his acts on camera. Collect and bag l... \n",
"\n",
" human_ref_B labels score_ratio \n",
"0 It’s ironic to me that research has shown that... 0 1.038462 \n",
"1 I'm pretty strict on time, to the point where ... 0 3.400000 \n",
"2 This sounds like a problem with a specific pro... 0 1.400000 \n",
"3 And when your teacher doesn't listen or pay at... 1 1.400000 \n",
"4 This sounds like a problem with a specific pro... 0 1.166667 \n",
"... ... ... ... \n",
"348713 Whatever you do, don't cut his trees down. 0 3.571429 \n",
"348714 Whatever you do, don't cut his trees down. 0 12.500000 \n",
"348715 Are you Rand Paul's neighbor? https://www.gq.... 1 1.285714 \n",
"348716 Just put up a fence. Legally he isn't responsi... 1 1.285714 \n",
"348717 If OP pays someone to clean his yard, and then... 1 3.500000 \n",
"\n",
"[348718 rows x 8 columns]"
]
},
"execution_count": 46,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# df['response_length'] = df['history'].apply(len)\n",
"# df['label'] = df['response_length'].apply(lambda x: 'long' if x > 100 else 'short')\n",
"df.drop(columns=['post_id', 'domain', 'c_root_id_A', 'c_root_id_B', 'created_at_utc_A', 'created_at_utc_B', 'seconds_difference'])"
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/riddhib/.pyenv/versions/3.10.13/lib/python3.10/site-packages/transformers/tokenization_utils_base.py:1617: FutureWarning: `clean_up_tokenization_spaces` was not set. It will be set to `True` by default. This behavior will be deprecated in transformers v4.45, and will be then set to `False` by default. For more details check this issue: https://github.com/huggingface/transformers/issues/31884\n",
" warnings.warn(\n"
]
}
],
"source": [
"model = AutoModelForCausalLMWithValueHead.from_pretrained(\"gpt2\")\n",
"ref_model = AutoModelForCausalLMWithValueHead.from_pretrained(\"gpt2\")\n",
"tokenizer = GPT2Tokenizer.from_pretrained(\"gpt2\")\n",
"tokenizer.pad_token = tokenizer.eos_token"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {},
"outputs": [],
"source": [
"from trl_rlhf_data import runner, ScriptArguments\n",
"import re\n",
"from dataclasses import dataclass\n",
"from typing import Dict, List, Optional\n",
"\n",
"from datasets import load_dataset\n",
"from transformers import HfArgumentParser"
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {},
"outputs": [
{
"ename": "TypeError",
"evalue": "runner() takes 0 positional arguments but 1 was given",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mTypeError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[0;32mIn[49], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m dataset \u001b[38;5;241m=\u001b[39m \u001b[43mrunner\u001b[49m\u001b[43m(\u001b[49m\u001b[43mScriptArguments\u001b[49m\u001b[43m)\u001b[49m\n",
"\u001b[0;31mTypeError\u001b[0m: runner() takes 0 positional arguments but 1 was given"
]
}
],
"source": [
"dataset = runner(ScriptArguments)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"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": "3.10.13"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
|