Update README.md
Browse files
README.md
CHANGED
|
@@ -1,491 +1,493 @@
|
|
| 1 |
-
---
|
| 2 |
-
dataset_info:
|
| 3 |
-
- config_name: C
|
| 4 |
-
features:
|
| 5 |
-
- name: blob_id
|
| 6 |
-
dtype: large_string
|
| 7 |
-
- name: language
|
| 8 |
-
dtype: large_string
|
| 9 |
-
- name: repo_name
|
| 10 |
-
dtype: large_string
|
| 11 |
-
- name: path
|
| 12 |
-
dtype: large_string
|
| 13 |
-
- name: src_encoding
|
| 14 |
-
dtype: large_string
|
| 15 |
-
- name: length_bytes
|
| 16 |
-
dtype: int64
|
| 17 |
-
- name: score
|
| 18 |
-
dtype: float64
|
| 19 |
-
- name: int_score
|
| 20 |
-
dtype: int64
|
| 21 |
-
- name: detected_licenses
|
| 22 |
-
large_list: large_string
|
| 23 |
-
- name: license_type
|
| 24 |
-
dtype: large_string
|
| 25 |
-
splits:
|
| 26 |
-
- name: train
|
| 27 |
-
num_bytes: 1100442974
|
| 28 |
-
num_examples: 5848375
|
| 29 |
-
download_size: 571816053
|
| 30 |
-
dataset_size: 1100442974
|
| 31 |
-
- config_name: CSharp
|
| 32 |
-
features:
|
| 33 |
-
- name: blob_id
|
| 34 |
-
dtype: large_string
|
| 35 |
-
- name: language
|
| 36 |
-
dtype: large_string
|
| 37 |
-
- name: repo_name
|
| 38 |
-
dtype: large_string
|
| 39 |
-
- name: path
|
| 40 |
-
dtype: large_string
|
| 41 |
-
- name: src_encoding
|
| 42 |
-
dtype: large_string
|
| 43 |
-
- name: length_bytes
|
| 44 |
-
dtype: int64
|
| 45 |
-
- name: score
|
| 46 |
-
dtype: float64
|
| 47 |
-
- name: int_score
|
| 48 |
-
dtype: int64
|
| 49 |
-
- name: detected_licenses
|
| 50 |
-
large_list: large_string
|
| 51 |
-
- name: license_type
|
| 52 |
-
dtype: large_string
|
| 53 |
-
splits:
|
| 54 |
-
- name: train
|
| 55 |
-
num_bytes: 2392066248
|
| 56 |
-
num_examples: 11425016
|
| 57 |
-
download_size: 1232015539
|
| 58 |
-
dataset_size: 2392066248
|
| 59 |
-
- config_name: Cpp
|
| 60 |
-
features:
|
| 61 |
-
- name: blob_id
|
| 62 |
-
dtype: large_string
|
| 63 |
-
- name: language
|
| 64 |
-
dtype: large_string
|
| 65 |
-
- name: repo_name
|
| 66 |
-
dtype: large_string
|
| 67 |
-
- name: path
|
| 68 |
-
dtype: large_string
|
| 69 |
-
- name: src_encoding
|
| 70 |
-
dtype: large_string
|
| 71 |
-
- name: length_bytes
|
| 72 |
-
dtype: int64
|
| 73 |
-
- name: score
|
| 74 |
-
dtype: float64
|
| 75 |
-
- name: int_score
|
| 76 |
-
dtype: int64
|
| 77 |
-
- name: detected_licenses
|
| 78 |
-
large_list: large_string
|
| 79 |
-
- name: license_type
|
| 80 |
-
dtype: large_string
|
| 81 |
-
splits:
|
| 82 |
-
- name: train
|
| 83 |
-
num_bytes: 3167426435
|
| 84 |
-
num_examples: 16246746
|
| 85 |
-
download_size: 1632803797
|
| 86 |
-
dataset_size: 3167426435
|
| 87 |
-
- config_name: Go
|
| 88 |
-
features:
|
| 89 |
-
- name: blob_id
|
| 90 |
-
dtype: large_string
|
| 91 |
-
- name: language
|
| 92 |
-
dtype: large_string
|
| 93 |
-
- name: repo_name
|
| 94 |
-
dtype: large_string
|
| 95 |
-
- name: path
|
| 96 |
-
dtype: large_string
|
| 97 |
-
- name: src_encoding
|
| 98 |
-
dtype: large_string
|
| 99 |
-
- name: length_bytes
|
| 100 |
-
dtype: int64
|
| 101 |
-
- name: score
|
| 102 |
-
dtype: float64
|
| 103 |
-
- name: int_score
|
| 104 |
-
dtype: int64
|
| 105 |
-
- name: detected_licenses
|
| 106 |
-
large_list: large_string
|
| 107 |
-
- name: license_type
|
| 108 |
-
dtype: large_string
|
| 109 |
-
- name: detected_licenses_right
|
| 110 |
-
large_list: large_string
|
| 111 |
-
- name: license_type_right
|
| 112 |
-
dtype: large_string
|
| 113 |
-
splits:
|
| 114 |
-
- name: train
|
| 115 |
-
num_bytes: 433053889
|
| 116 |
-
num_examples: 1917163
|
| 117 |
-
download_size: 179388495
|
| 118 |
-
dataset_size: 433053889
|
| 119 |
-
- config_name: Java
|
| 120 |
-
features:
|
| 121 |
-
- name: blob_id
|
| 122 |
-
dtype: large_string
|
| 123 |
-
- name: language
|
| 124 |
-
dtype: large_string
|
| 125 |
-
- name: repo_name
|
| 126 |
-
dtype: large_string
|
| 127 |
-
- name: path
|
| 128 |
-
dtype: large_string
|
| 129 |
-
- name: src_encoding
|
| 130 |
-
dtype: large_string
|
| 131 |
-
- name: length_bytes
|
| 132 |
-
dtype: int64
|
| 133 |
-
- name: score
|
| 134 |
-
dtype: float64
|
| 135 |
-
- name: int_score
|
| 136 |
-
dtype: int64
|
| 137 |
-
- name: detected_licenses
|
| 138 |
-
large_list: large_string
|
| 139 |
-
- name: license_type
|
| 140 |
-
dtype: large_string
|
| 141 |
-
splits:
|
| 142 |
-
- name: train
|
| 143 |
-
num_bytes: 10292427437
|
| 144 |
-
num_examples: 44990158
|
| 145 |
-
download_size: 5291667797
|
| 146 |
-
dataset_size: 10292427437
|
| 147 |
-
- config_name: JavaScript
|
| 148 |
-
features:
|
| 149 |
-
- name: blob_id
|
| 150 |
-
dtype: large_string
|
| 151 |
-
- name: language
|
| 152 |
-
dtype: large_string
|
| 153 |
-
- name: repo_name
|
| 154 |
-
dtype: large_string
|
| 155 |
-
- name: path
|
| 156 |
-
dtype: large_string
|
| 157 |
-
- name: src_encoding
|
| 158 |
-
dtype: large_string
|
| 159 |
-
- name: length_bytes
|
| 160 |
-
dtype: int64
|
| 161 |
-
- name: score
|
| 162 |
-
dtype: float64
|
| 163 |
-
- name: int_score
|
| 164 |
-
dtype: int64
|
| 165 |
-
- name: detected_licenses
|
| 166 |
-
large_list: large_string
|
| 167 |
-
- name: license_type
|
| 168 |
-
dtype: large_string
|
| 169 |
-
splits:
|
| 170 |
-
- name: train
|
| 171 |
-
num_bytes: 2654326008
|
| 172 |
-
num_examples: 13253431
|
| 173 |
-
download_size: 1287066511
|
| 174 |
-
dataset_size: 2654326008
|
| 175 |
-
- config_name: Markdown
|
| 176 |
-
features:
|
| 177 |
-
- name: blob_id
|
| 178 |
-
dtype: large_string
|
| 179 |
-
- name: language
|
| 180 |
-
dtype: large_string
|
| 181 |
-
- name: repo_name
|
| 182 |
-
dtype: large_string
|
| 183 |
-
- name: path
|
| 184 |
-
dtype: large_string
|
| 185 |
-
- name: src_encoding
|
| 186 |
-
dtype: large_string
|
| 187 |
-
- name: length_bytes
|
| 188 |
-
dtype: int64
|
| 189 |
-
- name: score
|
| 190 |
-
dtype: float64
|
| 191 |
-
- name: int_score
|
| 192 |
-
dtype: int64
|
| 193 |
-
- name: detected_licenses
|
| 194 |
-
large_list: large_string
|
| 195 |
-
- name: license_type
|
| 196 |
-
dtype: large_string
|
| 197 |
-
splits:
|
| 198 |
-
- name: train
|
| 199 |
-
num_bytes: 4268378053
|
| 200 |
-
num_examples: 20687077
|
| 201 |
-
download_size: 2058772192
|
| 202 |
-
dataset_size: 4268378053
|
| 203 |
-
- config_name: PHP
|
| 204 |
-
features:
|
| 205 |
-
- name: blob_id
|
| 206 |
-
dtype: large_string
|
| 207 |
-
- name: language
|
| 208 |
-
dtype: large_string
|
| 209 |
-
- name: repo_name
|
| 210 |
-
dtype: large_string
|
| 211 |
-
- name: path
|
| 212 |
-
dtype: large_string
|
| 213 |
-
- name: src_encoding
|
| 214 |
-
dtype: large_string
|
| 215 |
-
- name: length_bytes
|
| 216 |
-
dtype: int64
|
| 217 |
-
- name: score
|
| 218 |
-
dtype: float64
|
| 219 |
-
- name: int_score
|
| 220 |
-
dtype: int64
|
| 221 |
-
- name: detected_licenses
|
| 222 |
-
large_list: large_string
|
| 223 |
-
- name: license_type
|
| 224 |
-
dtype: large_string
|
| 225 |
-
splits:
|
| 226 |
-
- name: train
|
| 227 |
-
num_bytes: 1985843762
|
| 228 |
-
num_examples: 9914497
|
| 229 |
-
download_size: 983498806
|
| 230 |
-
dataset_size: 1985843762
|
| 231 |
-
- config_name: Python
|
| 232 |
-
features:
|
| 233 |
-
- name: blob_id
|
| 234 |
-
dtype: large_string
|
| 235 |
-
- name: language
|
| 236 |
-
dtype: large_string
|
| 237 |
-
- name: repo_name
|
| 238 |
-
dtype: large_string
|
| 239 |
-
- name: path
|
| 240 |
-
dtype: large_string
|
| 241 |
-
- name: src_encoding
|
| 242 |
-
dtype: large_string
|
| 243 |
-
- name: length_bytes
|
| 244 |
-
dtype: int64
|
| 245 |
-
- name: score
|
| 246 |
-
dtype: float64
|
| 247 |
-
- name: int_score
|
| 248 |
-
dtype: int64
|
| 249 |
-
- name: detected_licenses
|
| 250 |
-
large_list: large_string
|
| 251 |
-
- name: license_type
|
| 252 |
-
dtype: large_string
|
| 253 |
-
splits:
|
| 254 |
-
- name: train
|
| 255 |
-
num_bytes: 4947575770
|
| 256 |
-
num_examples: 25286019
|
| 257 |
-
download_size: 2500795086
|
| 258 |
-
dataset_size: 4947575770
|
| 259 |
-
- config_name: Ruby
|
| 260 |
-
features:
|
| 261 |
-
- name: blob_id
|
| 262 |
-
dtype: large_string
|
| 263 |
-
- name: language
|
| 264 |
-
dtype: large_string
|
| 265 |
-
- name: repo_name
|
| 266 |
-
dtype: large_string
|
| 267 |
-
- name: path
|
| 268 |
-
dtype: large_string
|
| 269 |
-
- name: src_encoding
|
| 270 |
-
dtype: large_string
|
| 271 |
-
- name: length_bytes
|
| 272 |
-
dtype: int64
|
| 273 |
-
- name: score
|
| 274 |
-
dtype: float64
|
| 275 |
-
- name: int_score
|
| 276 |
-
dtype: int64
|
| 277 |
-
- name: detected_licenses
|
| 278 |
-
large_list: large_string
|
| 279 |
-
- name: license_type
|
| 280 |
-
dtype: large_string
|
| 281 |
-
splits:
|
| 282 |
-
- name: train
|
| 283 |
-
num_bytes: 592832039
|
| 284 |
-
num_examples: 2976874
|
| 285 |
-
download_size: 284535771
|
| 286 |
-
dataset_size: 592832039
|
| 287 |
-
- config_name: Rust
|
| 288 |
-
features:
|
| 289 |
-
- name: blob_id
|
| 290 |
-
dtype: large_string
|
| 291 |
-
- name: language
|
| 292 |
-
dtype: large_string
|
| 293 |
-
- name: repo_name
|
| 294 |
-
dtype: large_string
|
| 295 |
-
- name: path
|
| 296 |
-
dtype: large_string
|
| 297 |
-
- name: src_encoding
|
| 298 |
-
dtype: large_string
|
| 299 |
-
- name: length_bytes
|
| 300 |
-
dtype: int64
|
| 301 |
-
- name: score
|
| 302 |
-
dtype: float64
|
| 303 |
-
- name: int_score
|
| 304 |
-
dtype: int64
|
| 305 |
-
- name: detected_licenses
|
| 306 |
-
large_list: large_string
|
| 307 |
-
- name: license_type
|
| 308 |
-
dtype: large_string
|
| 309 |
-
splits:
|
| 310 |
-
- name: train
|
| 311 |
-
num_bytes: 227434676
|
| 312 |
-
num_examples: 1135379
|
| 313 |
-
download_size: 103158397
|
| 314 |
-
dataset_size: 227434676
|
| 315 |
-
- config_name: SQL
|
| 316 |
-
features:
|
| 317 |
-
- name: blob_id
|
| 318 |
-
dtype: large_string
|
| 319 |
-
- name: language
|
| 320 |
-
dtype: large_string
|
| 321 |
-
- name: repo_name
|
| 322 |
-
dtype: large_string
|
| 323 |
-
- name: path
|
| 324 |
-
dtype: large_string
|
| 325 |
-
- name: src_encoding
|
| 326 |
-
dtype: large_string
|
| 327 |
-
- name: length_bytes
|
| 328 |
-
dtype: int64
|
| 329 |
-
- name: score
|
| 330 |
-
dtype: float64
|
| 331 |
-
- name: int_score
|
| 332 |
-
dtype: int64
|
| 333 |
-
- name: detected_licenses
|
| 334 |
-
large_list: large_string
|
| 335 |
-
- name: license_type
|
| 336 |
-
dtype: large_string
|
| 337 |
-
splits:
|
| 338 |
-
- name: train
|
| 339 |
-
num_bytes: 505669712
|
| 340 |
-
num_examples: 2504412
|
| 341 |
-
download_size: 261176608
|
| 342 |
-
dataset_size: 505669712
|
| 343 |
-
- config_name: Shell
|
| 344 |
-
features:
|
| 345 |
-
- name: blob_id
|
| 346 |
-
dtype: large_string
|
| 347 |
-
- name: language
|
| 348 |
-
dtype: large_string
|
| 349 |
-
- name: repo_name
|
| 350 |
-
dtype: large_string
|
| 351 |
-
- name: path
|
| 352 |
-
dtype: large_string
|
| 353 |
-
- name: src_encoding
|
| 354 |
-
dtype: large_string
|
| 355 |
-
- name: length_bytes
|
| 356 |
-
dtype: int64
|
| 357 |
-
- name: score
|
| 358 |
-
dtype: float64
|
| 359 |
-
- name: int_score
|
| 360 |
-
dtype: int64
|
| 361 |
-
- name: detected_licenses
|
| 362 |
-
large_list: large_string
|
| 363 |
-
- name: license_type
|
| 364 |
-
dtype: large_string
|
| 365 |
-
splits:
|
| 366 |
-
- name: train
|
| 367 |
-
num_bytes: 811611733
|
| 368 |
-
num_examples: 4133547
|
| 369 |
-
download_size: 394872047
|
| 370 |
-
dataset_size: 811611733
|
| 371 |
-
- config_name: Swift
|
| 372 |
-
features:
|
| 373 |
-
- name: blob_id
|
| 374 |
-
dtype: large_string
|
| 375 |
-
- name: language
|
| 376 |
-
dtype: large_string
|
| 377 |
-
- name: repo_name
|
| 378 |
-
dtype: large_string
|
| 379 |
-
- name: path
|
| 380 |
-
dtype: large_string
|
| 381 |
-
- name: src_encoding
|
| 382 |
-
dtype: large_string
|
| 383 |
-
- name: length_bytes
|
| 384 |
-
dtype: int64
|
| 385 |
-
- name: score
|
| 386 |
-
dtype: float64
|
| 387 |
-
- name: int_score
|
| 388 |
-
dtype: int64
|
| 389 |
-
- name: detected_licenses
|
| 390 |
-
large_list: large_string
|
| 391 |
-
- name: license_type
|
| 392 |
-
dtype: large_string
|
| 393 |
-
splits:
|
| 394 |
-
- name: train
|
| 395 |
-
num_bytes: 529873695
|
| 396 |
-
num_examples: 2454309
|
| 397 |
-
download_size: 257883733
|
| 398 |
-
dataset_size: 529873695
|
| 399 |
-
- config_name: TypeScript
|
| 400 |
-
features:
|
| 401 |
-
- name: blob_id
|
| 402 |
-
dtype: large_string
|
| 403 |
-
- name: language
|
| 404 |
-
dtype: large_string
|
| 405 |
-
- name: repo_name
|
| 406 |
-
dtype: large_string
|
| 407 |
-
- name: path
|
| 408 |
-
dtype: large_string
|
| 409 |
-
- name: src_encoding
|
| 410 |
-
dtype: large_string
|
| 411 |
-
- name: length_bytes
|
| 412 |
-
dtype: int64
|
| 413 |
-
- name: score
|
| 414 |
-
dtype: float64
|
| 415 |
-
- name: int_score
|
| 416 |
-
dtype: int64
|
| 417 |
-
- name: detected_licenses
|
| 418 |
-
large_list: large_string
|
| 419 |
-
- name: license_type
|
| 420 |
-
dtype: large_string
|
| 421 |
-
splits:
|
| 422 |
-
- name: train
|
| 423 |
-
num_bytes: 904736029
|
| 424 |
-
num_examples: 4290356
|
| 425 |
-
download_size: 425942502
|
| 426 |
-
dataset_size: 904736029
|
| 427 |
-
configs:
|
| 428 |
-
- config_name: C
|
| 429 |
-
data_files:
|
| 430 |
-
- split: train
|
| 431 |
-
path: C/train-*
|
| 432 |
-
- config_name: CSharp
|
| 433 |
-
data_files:
|
| 434 |
-
- split: train
|
| 435 |
-
path: CSharp/train-*
|
| 436 |
-
- config_name: Cpp
|
| 437 |
-
data_files:
|
| 438 |
-
- split: train
|
| 439 |
-
path: Cpp/train-*
|
| 440 |
-
- config_name: Go
|
| 441 |
-
data_files:
|
| 442 |
-
- split: train
|
| 443 |
-
path: Go/train-*
|
| 444 |
-
- config_name: Java
|
| 445 |
-
data_files:
|
| 446 |
-
- split: train
|
| 447 |
-
path: Java/train-*
|
| 448 |
-
- config_name: JavaScript
|
| 449 |
-
data_files:
|
| 450 |
-
- split: train
|
| 451 |
-
path: JavaScript/train-*
|
| 452 |
-
- config_name: Markdown
|
| 453 |
-
data_files:
|
| 454 |
-
- split: train
|
| 455 |
-
path: Markdown/train-*
|
| 456 |
-
- config_name: PHP
|
| 457 |
-
data_files:
|
| 458 |
-
- split: train
|
| 459 |
-
path: PHP/train-*
|
| 460 |
-
- config_name: Python
|
| 461 |
-
data_files:
|
| 462 |
-
- split: train
|
| 463 |
-
path: Python/train-*
|
| 464 |
-
- config_name: Ruby
|
| 465 |
-
data_files:
|
| 466 |
-
- split: train
|
| 467 |
-
path: Ruby/train-*
|
| 468 |
-
- config_name: Rust
|
| 469 |
-
data_files:
|
| 470 |
-
- split: train
|
| 471 |
-
path: Rust/train-*
|
| 472 |
-
- config_name: SQL
|
| 473 |
-
data_files:
|
| 474 |
-
- split: train
|
| 475 |
-
path: SQL/train-*
|
| 476 |
-
- config_name: Shell
|
| 477 |
-
data_files:
|
| 478 |
-
- split: train
|
| 479 |
-
path: Shell/train-*
|
| 480 |
-
- config_name: Swift
|
| 481 |
-
data_files:
|
| 482 |
-
- split: train
|
| 483 |
-
path: Swift/train-*
|
| 484 |
-
- config_name: TypeScript
|
| 485 |
-
data_files:
|
| 486 |
-
- split: train
|
| 487 |
-
path: TypeScript/train-*
|
| 488 |
-
|
|
|
|
|
|
|
| 489 |
|
| 490 |
[HuggingFaceTB/stack-edu](https://huggingface.co/datasets/HuggingFaceTB/stack-edu) 에서 한국어가 포함된 데이터만 발췌
|
| 491 |
|
|
@@ -519,87 +521,7 @@ configs:
|
|
| 519 |
|
| 520 |
|
| 521 |
|
| 522 |
-
|
| 523 |
-
# 💻 Stack-Edu
|
| 524 |
-
|
| 525 |
-
|
| 526 |
-
|
| 527 |
-
Stack-Edu is a 125B token dataset of educational code filtered from [The Stack v2](https://huggingface.co/datasets/bigcode/the-stack-v2), precisely the curated training corpus of [StarCoder2](https://arxiv.org/abs/2402.19173) models denoted StarCoder2Data. It is intended for Language Models training.
|
| 528 |
-
|
| 529 |
-
This dataset was curated using a classifier-based filtering strategy, inspired by [📚 FineWeb-Edu](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu), to retain only the highest-quality educational programming content.
|
| 530 |
-
|
| 531 |
-
Stack-Edu shows consistent improvement over StarCoder2data on all the programming languages on MultiPL-E benchmark.
|
| 532 |
-
|
| 533 |
-
<img src="https://cdn-uploads.huggingface.co/production/uploads/61c141342aac764ce1654e43/GWnPgD0diMu0I8buK6mvG.png" width="600"/>
|
| 534 |
-
|
| 535 |
-
## Downloading the data
|
| 536 |
-
This dataset only contains the SWHIDs to download the code files and not the content of the files itself. The contents can be downloaded from Software Heritage's S3 bucket to ensure data compliance.
|
| 537 |
-
Please refer to [the-stack-v2](https://huggingface.co/datasets/bigcode/the-stack-v2-train-full-ids) for the data license.
|
| 538 |
-
|
| 539 |
-
When running on a 16-core AWS `us-east-1` instance, this script takes ~6 hours to download the files:
|
| 540 |
-
```python
|
| 541 |
-
import boto3
|
| 542 |
-
import gzip
|
| 543 |
-
from datasets import load_dataset
|
| 544 |
-
from botocore.exceptions import ClientError
|
| 545 |
-
|
| 546 |
-
num_proc = 16
|
| 547 |
-
s3 = boto3.client('s3')
|
| 548 |
-
bucket_name = "softwareheritage"
|
| 549 |
-
|
| 550 |
-
def download_contents(blob_id):
|
| 551 |
-
key = f"content/{blob_id}"
|
| 552 |
-
try:
|
| 553 |
-
obj = s3.get_object(Bucket=bucket_name, Key=key)
|
| 554 |
-
with gzip.GzipFile(fileobj=obj['Body']) as fin:
|
| 555 |
-
content = fin.read().decode("utf-8", errors="ignore")
|
| 556 |
-
return {"text": content, "download_success": True}
|
| 557 |
-
except ClientError as e:
|
| 558 |
-
if e.response['Error']['Code'] == 'NoSuchKey':
|
| 559 |
-
print(f"File not found: {key}")
|
| 560 |
-
return {"text": "", "download_success": False}
|
| 561 |
-
else:
|
| 562 |
-
raise
|
| 563 |
-
|
| 564 |
-
# For Python
|
| 565 |
-
ds = load_dataset("HuggingFaceTB/stack-edu", "Python", split="train", num_proc=num_proc)
|
| 566 |
-
ds = ds.map(download_contents, input_columns="blob_id", num_proc=num_proc)
|
| 567 |
-
|
| 568 |
-
# Filter out failed downloads
|
| 569 |
-
ds = ds.filter(lambda x: x['download_success'])
|
| 570 |
-
|
| 571 |
-
# Optionally, print the first example to verify the data
|
| 572 |
-
print(ds[0])
|
| 573 |
-
```
|
| 574 |
-
|
| 575 |
-
## Details
|
| 576 |
-
The table below shows the number of tokens in each programming language using [SmolLM2](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B-Instruct) tokenizer.
|
| 577 |
-
|
| 578 |
-
| Language | Stack-Edu (B tokens) |
|
| 579 |
-
|------------|----------------------|
|
| 580 |
-
| Python | 21.8 |
|
| 581 |
-
| Cpp | 16.0 |
|
| 582 |
-
| Markdown | 14.0 |
|
| 583 |
-
| C | 11.1 |
|
| 584 |
-
| JavaScript | 11.1 |
|
| 585 |
-
| Java | 42.1 |
|
| 586 |
-
| SQL | 9.62 |
|
| 587 |
-
| PHP | 9.07 |
|
| 588 |
-
| C-Sharp | 8.87 |
|
| 589 |
-
| TypeScript | 3.03 |
|
| 590 |
-
| Shell | 3.13 |
|
| 591 |
-
| Swift | 1.83 |
|
| 592 |
-
| Go | 1.80 |
|
| 593 |
-
| Rust | 1.75 |
|
| 594 |
-
| Ruby | 1.61 |
|
| 595 |
-
|
| 596 |
-
|
| 597 |
-
## Dataset curation
|
| 598 |
-
To build Stack-Edu, we:
|
| 599 |
-
|
| 600 |
-
- Selected 15 largest programming languages from StarCoder2Data
|
| 601 |
-
- Trained 15 language-specific classifiers, using [StarEncoder](https://huggingface.co/bigcode/starencoder) model on synthetic annotations generated by Llama3-70B-Instruct. The classifiers for each language are available in this [collection](https://huggingface.co/collections/HuggingFaceTB/the-ultimate-collection-of-code-classifiers-67b5aa3eb8994a4b71453005).
|
| 602 |
-
- Applied a filtering threshold of 3 (out of 5) to retain highly educational content, except for Java, which performed best at threshold 2.
|
| 603 |
|
| 604 |
## Citation Information
|
| 605 |
|
|
|
|
| 1 |
+
---
|
| 2 |
+
dataset_info:
|
| 3 |
+
- config_name: C
|
| 4 |
+
features:
|
| 5 |
+
- name: blob_id
|
| 6 |
+
dtype: large_string
|
| 7 |
+
- name: language
|
| 8 |
+
dtype: large_string
|
| 9 |
+
- name: repo_name
|
| 10 |
+
dtype: large_string
|
| 11 |
+
- name: path
|
| 12 |
+
dtype: large_string
|
| 13 |
+
- name: src_encoding
|
| 14 |
+
dtype: large_string
|
| 15 |
+
- name: length_bytes
|
| 16 |
+
dtype: int64
|
| 17 |
+
- name: score
|
| 18 |
+
dtype: float64
|
| 19 |
+
- name: int_score
|
| 20 |
+
dtype: int64
|
| 21 |
+
- name: detected_licenses
|
| 22 |
+
large_list: large_string
|
| 23 |
+
- name: license_type
|
| 24 |
+
dtype: large_string
|
| 25 |
+
splits:
|
| 26 |
+
- name: train
|
| 27 |
+
num_bytes: 1100442974
|
| 28 |
+
num_examples: 5848375
|
| 29 |
+
download_size: 571816053
|
| 30 |
+
dataset_size: 1100442974
|
| 31 |
+
- config_name: CSharp
|
| 32 |
+
features:
|
| 33 |
+
- name: blob_id
|
| 34 |
+
dtype: large_string
|
| 35 |
+
- name: language
|
| 36 |
+
dtype: large_string
|
| 37 |
+
- name: repo_name
|
| 38 |
+
dtype: large_string
|
| 39 |
+
- name: path
|
| 40 |
+
dtype: large_string
|
| 41 |
+
- name: src_encoding
|
| 42 |
+
dtype: large_string
|
| 43 |
+
- name: length_bytes
|
| 44 |
+
dtype: int64
|
| 45 |
+
- name: score
|
| 46 |
+
dtype: float64
|
| 47 |
+
- name: int_score
|
| 48 |
+
dtype: int64
|
| 49 |
+
- name: detected_licenses
|
| 50 |
+
large_list: large_string
|
| 51 |
+
- name: license_type
|
| 52 |
+
dtype: large_string
|
| 53 |
+
splits:
|
| 54 |
+
- name: train
|
| 55 |
+
num_bytes: 2392066248
|
| 56 |
+
num_examples: 11425016
|
| 57 |
+
download_size: 1232015539
|
| 58 |
+
dataset_size: 2392066248
|
| 59 |
+
- config_name: Cpp
|
| 60 |
+
features:
|
| 61 |
+
- name: blob_id
|
| 62 |
+
dtype: large_string
|
| 63 |
+
- name: language
|
| 64 |
+
dtype: large_string
|
| 65 |
+
- name: repo_name
|
| 66 |
+
dtype: large_string
|
| 67 |
+
- name: path
|
| 68 |
+
dtype: large_string
|
| 69 |
+
- name: src_encoding
|
| 70 |
+
dtype: large_string
|
| 71 |
+
- name: length_bytes
|
| 72 |
+
dtype: int64
|
| 73 |
+
- name: score
|
| 74 |
+
dtype: float64
|
| 75 |
+
- name: int_score
|
| 76 |
+
dtype: int64
|
| 77 |
+
- name: detected_licenses
|
| 78 |
+
large_list: large_string
|
| 79 |
+
- name: license_type
|
| 80 |
+
dtype: large_string
|
| 81 |
+
splits:
|
| 82 |
+
- name: train
|
| 83 |
+
num_bytes: 3167426435
|
| 84 |
+
num_examples: 16246746
|
| 85 |
+
download_size: 1632803797
|
| 86 |
+
dataset_size: 3167426435
|
| 87 |
+
- config_name: Go
|
| 88 |
+
features:
|
| 89 |
+
- name: blob_id
|
| 90 |
+
dtype: large_string
|
| 91 |
+
- name: language
|
| 92 |
+
dtype: large_string
|
| 93 |
+
- name: repo_name
|
| 94 |
+
dtype: large_string
|
| 95 |
+
- name: path
|
| 96 |
+
dtype: large_string
|
| 97 |
+
- name: src_encoding
|
| 98 |
+
dtype: large_string
|
| 99 |
+
- name: length_bytes
|
| 100 |
+
dtype: int64
|
| 101 |
+
- name: score
|
| 102 |
+
dtype: float64
|
| 103 |
+
- name: int_score
|
| 104 |
+
dtype: int64
|
| 105 |
+
- name: detected_licenses
|
| 106 |
+
large_list: large_string
|
| 107 |
+
- name: license_type
|
| 108 |
+
dtype: large_string
|
| 109 |
+
- name: detected_licenses_right
|
| 110 |
+
large_list: large_string
|
| 111 |
+
- name: license_type_right
|
| 112 |
+
dtype: large_string
|
| 113 |
+
splits:
|
| 114 |
+
- name: train
|
| 115 |
+
num_bytes: 433053889
|
| 116 |
+
num_examples: 1917163
|
| 117 |
+
download_size: 179388495
|
| 118 |
+
dataset_size: 433053889
|
| 119 |
+
- config_name: Java
|
| 120 |
+
features:
|
| 121 |
+
- name: blob_id
|
| 122 |
+
dtype: large_string
|
| 123 |
+
- name: language
|
| 124 |
+
dtype: large_string
|
| 125 |
+
- name: repo_name
|
| 126 |
+
dtype: large_string
|
| 127 |
+
- name: path
|
| 128 |
+
dtype: large_string
|
| 129 |
+
- name: src_encoding
|
| 130 |
+
dtype: large_string
|
| 131 |
+
- name: length_bytes
|
| 132 |
+
dtype: int64
|
| 133 |
+
- name: score
|
| 134 |
+
dtype: float64
|
| 135 |
+
- name: int_score
|
| 136 |
+
dtype: int64
|
| 137 |
+
- name: detected_licenses
|
| 138 |
+
large_list: large_string
|
| 139 |
+
- name: license_type
|
| 140 |
+
dtype: large_string
|
| 141 |
+
splits:
|
| 142 |
+
- name: train
|
| 143 |
+
num_bytes: 10292427437
|
| 144 |
+
num_examples: 44990158
|
| 145 |
+
download_size: 5291667797
|
| 146 |
+
dataset_size: 10292427437
|
| 147 |
+
- config_name: JavaScript
|
| 148 |
+
features:
|
| 149 |
+
- name: blob_id
|
| 150 |
+
dtype: large_string
|
| 151 |
+
- name: language
|
| 152 |
+
dtype: large_string
|
| 153 |
+
- name: repo_name
|
| 154 |
+
dtype: large_string
|
| 155 |
+
- name: path
|
| 156 |
+
dtype: large_string
|
| 157 |
+
- name: src_encoding
|
| 158 |
+
dtype: large_string
|
| 159 |
+
- name: length_bytes
|
| 160 |
+
dtype: int64
|
| 161 |
+
- name: score
|
| 162 |
+
dtype: float64
|
| 163 |
+
- name: int_score
|
| 164 |
+
dtype: int64
|
| 165 |
+
- name: detected_licenses
|
| 166 |
+
large_list: large_string
|
| 167 |
+
- name: license_type
|
| 168 |
+
dtype: large_string
|
| 169 |
+
splits:
|
| 170 |
+
- name: train
|
| 171 |
+
num_bytes: 2654326008
|
| 172 |
+
num_examples: 13253431
|
| 173 |
+
download_size: 1287066511
|
| 174 |
+
dataset_size: 2654326008
|
| 175 |
+
- config_name: Markdown
|
| 176 |
+
features:
|
| 177 |
+
- name: blob_id
|
| 178 |
+
dtype: large_string
|
| 179 |
+
- name: language
|
| 180 |
+
dtype: large_string
|
| 181 |
+
- name: repo_name
|
| 182 |
+
dtype: large_string
|
| 183 |
+
- name: path
|
| 184 |
+
dtype: large_string
|
| 185 |
+
- name: src_encoding
|
| 186 |
+
dtype: large_string
|
| 187 |
+
- name: length_bytes
|
| 188 |
+
dtype: int64
|
| 189 |
+
- name: score
|
| 190 |
+
dtype: float64
|
| 191 |
+
- name: int_score
|
| 192 |
+
dtype: int64
|
| 193 |
+
- name: detected_licenses
|
| 194 |
+
large_list: large_string
|
| 195 |
+
- name: license_type
|
| 196 |
+
dtype: large_string
|
| 197 |
+
splits:
|
| 198 |
+
- name: train
|
| 199 |
+
num_bytes: 4268378053
|
| 200 |
+
num_examples: 20687077
|
| 201 |
+
download_size: 2058772192
|
| 202 |
+
dataset_size: 4268378053
|
| 203 |
+
- config_name: PHP
|
| 204 |
+
features:
|
| 205 |
+
- name: blob_id
|
| 206 |
+
dtype: large_string
|
| 207 |
+
- name: language
|
| 208 |
+
dtype: large_string
|
| 209 |
+
- name: repo_name
|
| 210 |
+
dtype: large_string
|
| 211 |
+
- name: path
|
| 212 |
+
dtype: large_string
|
| 213 |
+
- name: src_encoding
|
| 214 |
+
dtype: large_string
|
| 215 |
+
- name: length_bytes
|
| 216 |
+
dtype: int64
|
| 217 |
+
- name: score
|
| 218 |
+
dtype: float64
|
| 219 |
+
- name: int_score
|
| 220 |
+
dtype: int64
|
| 221 |
+
- name: detected_licenses
|
| 222 |
+
large_list: large_string
|
| 223 |
+
- name: license_type
|
| 224 |
+
dtype: large_string
|
| 225 |
+
splits:
|
| 226 |
+
- name: train
|
| 227 |
+
num_bytes: 1985843762
|
| 228 |
+
num_examples: 9914497
|
| 229 |
+
download_size: 983498806
|
| 230 |
+
dataset_size: 1985843762
|
| 231 |
+
- config_name: Python
|
| 232 |
+
features:
|
| 233 |
+
- name: blob_id
|
| 234 |
+
dtype: large_string
|
| 235 |
+
- name: language
|
| 236 |
+
dtype: large_string
|
| 237 |
+
- name: repo_name
|
| 238 |
+
dtype: large_string
|
| 239 |
+
- name: path
|
| 240 |
+
dtype: large_string
|
| 241 |
+
- name: src_encoding
|
| 242 |
+
dtype: large_string
|
| 243 |
+
- name: length_bytes
|
| 244 |
+
dtype: int64
|
| 245 |
+
- name: score
|
| 246 |
+
dtype: float64
|
| 247 |
+
- name: int_score
|
| 248 |
+
dtype: int64
|
| 249 |
+
- name: detected_licenses
|
| 250 |
+
large_list: large_string
|
| 251 |
+
- name: license_type
|
| 252 |
+
dtype: large_string
|
| 253 |
+
splits:
|
| 254 |
+
- name: train
|
| 255 |
+
num_bytes: 4947575770
|
| 256 |
+
num_examples: 25286019
|
| 257 |
+
download_size: 2500795086
|
| 258 |
+
dataset_size: 4947575770
|
| 259 |
+
- config_name: Ruby
|
| 260 |
+
features:
|
| 261 |
+
- name: blob_id
|
| 262 |
+
dtype: large_string
|
| 263 |
+
- name: language
|
| 264 |
+
dtype: large_string
|
| 265 |
+
- name: repo_name
|
| 266 |
+
dtype: large_string
|
| 267 |
+
- name: path
|
| 268 |
+
dtype: large_string
|
| 269 |
+
- name: src_encoding
|
| 270 |
+
dtype: large_string
|
| 271 |
+
- name: length_bytes
|
| 272 |
+
dtype: int64
|
| 273 |
+
- name: score
|
| 274 |
+
dtype: float64
|
| 275 |
+
- name: int_score
|
| 276 |
+
dtype: int64
|
| 277 |
+
- name: detected_licenses
|
| 278 |
+
large_list: large_string
|
| 279 |
+
- name: license_type
|
| 280 |
+
dtype: large_string
|
| 281 |
+
splits:
|
| 282 |
+
- name: train
|
| 283 |
+
num_bytes: 592832039
|
| 284 |
+
num_examples: 2976874
|
| 285 |
+
download_size: 284535771
|
| 286 |
+
dataset_size: 592832039
|
| 287 |
+
- config_name: Rust
|
| 288 |
+
features:
|
| 289 |
+
- name: blob_id
|
| 290 |
+
dtype: large_string
|
| 291 |
+
- name: language
|
| 292 |
+
dtype: large_string
|
| 293 |
+
- name: repo_name
|
| 294 |
+
dtype: large_string
|
| 295 |
+
- name: path
|
| 296 |
+
dtype: large_string
|
| 297 |
+
- name: src_encoding
|
| 298 |
+
dtype: large_string
|
| 299 |
+
- name: length_bytes
|
| 300 |
+
dtype: int64
|
| 301 |
+
- name: score
|
| 302 |
+
dtype: float64
|
| 303 |
+
- name: int_score
|
| 304 |
+
dtype: int64
|
| 305 |
+
- name: detected_licenses
|
| 306 |
+
large_list: large_string
|
| 307 |
+
- name: license_type
|
| 308 |
+
dtype: large_string
|
| 309 |
+
splits:
|
| 310 |
+
- name: train
|
| 311 |
+
num_bytes: 227434676
|
| 312 |
+
num_examples: 1135379
|
| 313 |
+
download_size: 103158397
|
| 314 |
+
dataset_size: 227434676
|
| 315 |
+
- config_name: SQL
|
| 316 |
+
features:
|
| 317 |
+
- name: blob_id
|
| 318 |
+
dtype: large_string
|
| 319 |
+
- name: language
|
| 320 |
+
dtype: large_string
|
| 321 |
+
- name: repo_name
|
| 322 |
+
dtype: large_string
|
| 323 |
+
- name: path
|
| 324 |
+
dtype: large_string
|
| 325 |
+
- name: src_encoding
|
| 326 |
+
dtype: large_string
|
| 327 |
+
- name: length_bytes
|
| 328 |
+
dtype: int64
|
| 329 |
+
- name: score
|
| 330 |
+
dtype: float64
|
| 331 |
+
- name: int_score
|
| 332 |
+
dtype: int64
|
| 333 |
+
- name: detected_licenses
|
| 334 |
+
large_list: large_string
|
| 335 |
+
- name: license_type
|
| 336 |
+
dtype: large_string
|
| 337 |
+
splits:
|
| 338 |
+
- name: train
|
| 339 |
+
num_bytes: 505669712
|
| 340 |
+
num_examples: 2504412
|
| 341 |
+
download_size: 261176608
|
| 342 |
+
dataset_size: 505669712
|
| 343 |
+
- config_name: Shell
|
| 344 |
+
features:
|
| 345 |
+
- name: blob_id
|
| 346 |
+
dtype: large_string
|
| 347 |
+
- name: language
|
| 348 |
+
dtype: large_string
|
| 349 |
+
- name: repo_name
|
| 350 |
+
dtype: large_string
|
| 351 |
+
- name: path
|
| 352 |
+
dtype: large_string
|
| 353 |
+
- name: src_encoding
|
| 354 |
+
dtype: large_string
|
| 355 |
+
- name: length_bytes
|
| 356 |
+
dtype: int64
|
| 357 |
+
- name: score
|
| 358 |
+
dtype: float64
|
| 359 |
+
- name: int_score
|
| 360 |
+
dtype: int64
|
| 361 |
+
- name: detected_licenses
|
| 362 |
+
large_list: large_string
|
| 363 |
+
- name: license_type
|
| 364 |
+
dtype: large_string
|
| 365 |
+
splits:
|
| 366 |
+
- name: train
|
| 367 |
+
num_bytes: 811611733
|
| 368 |
+
num_examples: 4133547
|
| 369 |
+
download_size: 394872047
|
| 370 |
+
dataset_size: 811611733
|
| 371 |
+
- config_name: Swift
|
| 372 |
+
features:
|
| 373 |
+
- name: blob_id
|
| 374 |
+
dtype: large_string
|
| 375 |
+
- name: language
|
| 376 |
+
dtype: large_string
|
| 377 |
+
- name: repo_name
|
| 378 |
+
dtype: large_string
|
| 379 |
+
- name: path
|
| 380 |
+
dtype: large_string
|
| 381 |
+
- name: src_encoding
|
| 382 |
+
dtype: large_string
|
| 383 |
+
- name: length_bytes
|
| 384 |
+
dtype: int64
|
| 385 |
+
- name: score
|
| 386 |
+
dtype: float64
|
| 387 |
+
- name: int_score
|
| 388 |
+
dtype: int64
|
| 389 |
+
- name: detected_licenses
|
| 390 |
+
large_list: large_string
|
| 391 |
+
- name: license_type
|
| 392 |
+
dtype: large_string
|
| 393 |
+
splits:
|
| 394 |
+
- name: train
|
| 395 |
+
num_bytes: 529873695
|
| 396 |
+
num_examples: 2454309
|
| 397 |
+
download_size: 257883733
|
| 398 |
+
dataset_size: 529873695
|
| 399 |
+
- config_name: TypeScript
|
| 400 |
+
features:
|
| 401 |
+
- name: blob_id
|
| 402 |
+
dtype: large_string
|
| 403 |
+
- name: language
|
| 404 |
+
dtype: large_string
|
| 405 |
+
- name: repo_name
|
| 406 |
+
dtype: large_string
|
| 407 |
+
- name: path
|
| 408 |
+
dtype: large_string
|
| 409 |
+
- name: src_encoding
|
| 410 |
+
dtype: large_string
|
| 411 |
+
- name: length_bytes
|
| 412 |
+
dtype: int64
|
| 413 |
+
- name: score
|
| 414 |
+
dtype: float64
|
| 415 |
+
- name: int_score
|
| 416 |
+
dtype: int64
|
| 417 |
+
- name: detected_licenses
|
| 418 |
+
large_list: large_string
|
| 419 |
+
- name: license_type
|
| 420 |
+
dtype: large_string
|
| 421 |
+
splits:
|
| 422 |
+
- name: train
|
| 423 |
+
num_bytes: 904736029
|
| 424 |
+
num_examples: 4290356
|
| 425 |
+
download_size: 425942502
|
| 426 |
+
dataset_size: 904736029
|
| 427 |
+
configs:
|
| 428 |
+
- config_name: C
|
| 429 |
+
data_files:
|
| 430 |
+
- split: train
|
| 431 |
+
path: C/train-*
|
| 432 |
+
- config_name: CSharp
|
| 433 |
+
data_files:
|
| 434 |
+
- split: train
|
| 435 |
+
path: CSharp/train-*
|
| 436 |
+
- config_name: Cpp
|
| 437 |
+
data_files:
|
| 438 |
+
- split: train
|
| 439 |
+
path: Cpp/train-*
|
| 440 |
+
- config_name: Go
|
| 441 |
+
data_files:
|
| 442 |
+
- split: train
|
| 443 |
+
path: Go/train-*
|
| 444 |
+
- config_name: Java
|
| 445 |
+
data_files:
|
| 446 |
+
- split: train
|
| 447 |
+
path: Java/train-*
|
| 448 |
+
- config_name: JavaScript
|
| 449 |
+
data_files:
|
| 450 |
+
- split: train
|
| 451 |
+
path: JavaScript/train-*
|
| 452 |
+
- config_name: Markdown
|
| 453 |
+
data_files:
|
| 454 |
+
- split: train
|
| 455 |
+
path: Markdown/train-*
|
| 456 |
+
- config_name: PHP
|
| 457 |
+
data_files:
|
| 458 |
+
- split: train
|
| 459 |
+
path: PHP/train-*
|
| 460 |
+
- config_name: Python
|
| 461 |
+
data_files:
|
| 462 |
+
- split: train
|
| 463 |
+
path: Python/train-*
|
| 464 |
+
- config_name: Ruby
|
| 465 |
+
data_files:
|
| 466 |
+
- split: train
|
| 467 |
+
path: Ruby/train-*
|
| 468 |
+
- config_name: Rust
|
| 469 |
+
data_files:
|
| 470 |
+
- split: train
|
| 471 |
+
path: Rust/train-*
|
| 472 |
+
- config_name: SQL
|
| 473 |
+
data_files:
|
| 474 |
+
- split: train
|
| 475 |
+
path: SQL/train-*
|
| 476 |
+
- config_name: Shell
|
| 477 |
+
data_files:
|
| 478 |
+
- split: train
|
| 479 |
+
path: Shell/train-*
|
| 480 |
+
- config_name: Swift
|
| 481 |
+
data_files:
|
| 482 |
+
- split: train
|
| 483 |
+
path: Swift/train-*
|
| 484 |
+
- config_name: TypeScript
|
| 485 |
+
data_files:
|
| 486 |
+
- split: train
|
| 487 |
+
path: TypeScript/train-*
|
| 488 |
+
language:
|
| 489 |
+
- ko
|
| 490 |
+
---
|
| 491 |
|
| 492 |
[HuggingFaceTB/stack-edu](https://huggingface.co/datasets/HuggingFaceTB/stack-edu) 에서 한국어가 포함된 데이터만 발췌
|
| 493 |
|
|
|
|
| 521 |
|
| 522 |
|
| 523 |
|
| 524 |
+
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 525 |
|
| 526 |
## Citation Information
|
| 527 |
|