unitorch.cli.models.blip¤
BlipProcessor¤
Tip
core/process/blip
is the section for configuration of BlipProcessor.
Bases: BlipProcessor
Processor for the BLIP model.
Initialize BlipProcessor.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
vocab_path |
str
|
The path to the vocabulary file. |
required |
vision_config_path |
str
|
The path to the vision configuration file. |
required |
max_seq_length |
int
|
The maximum sequence length. Defaults to 128. |
128
|
max_gen_seq_length |
int
|
The maximum generated sequence length. Defaults to 48. |
48
|
position_start_id |
int
|
The start position ID. Defaults to 0. |
0
|
Source code in src/unitorch/cli/models/blip/processing.py
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 |
|
from_core_configure
classmethod
¤
from_core_configure(config, **kwargs)
Create an instance of BlipProcessor from a core configuration.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config |
The core configuration. |
required | |
**kwargs |
Additional keyword arguments. |
{}
|
Returns:
Name | Type | Description |
---|---|---|
dict |
A dictionary containing the processor parameters. |
Source code in src/unitorch/cli/models/blip/processing.py
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 |
|
BlipForPretrain¤
Tip
core/model/pretrain/blip
is the section for configuration of BlipForPretrain.
Bases: BlipForPretrain
BLIP model for pretraining.
Initialize BlipForPretrain.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config_path |
str
|
The path to the model configuration file. |
required |
projection_dim |
int
|
The dimension of the projection head. Defaults to 512. |
512
|
freeze_base_model |
bool
|
Whether to freeze the base model parameters. Defaults to True. |
True
|
gradient_checkpointing |
bool
|
Whether to use gradient checkpointing. Defaults to False. |
False
|
use_all_gather |
bool
|
Whether to use all_gather operation. Defaults to True. |
True
|
Source code in src/unitorch/cli/models/blip/modeling.py
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 |
|
forward ¤
forward(
input_ids: Tensor,
pixel_values: Tensor,
attention_mask: Tensor = None,
position_ids: Tensor = None,
)
Forward pass of the BlipForPretrain model.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
input_ids |
Tensor
|
The input token IDs. |
required |
pixel_values |
Tensor
|
The pixel values of the images. |
required |
attention_mask |
Tensor
|
The attention mask. Defaults to None. |
None
|
position_ids |
Tensor
|
The position IDs. Defaults to None. |
None
|
Returns:
Name | Type | Description |
---|---|---|
LossOutputs |
The loss outputs. |
Source code in src/unitorch/cli/models/blip/modeling.py
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 |
|
from_core_configure
classmethod
¤
from_core_configure(config, **kwargs)
Create an instance of BlipForPretrain from a core configuration.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config |
The core configuration. |
required | |
**kwargs |
Additional keyword arguments. |
{}
|
Returns:
Name | Type | Description |
---|---|---|
BlipForPretrain |
An instance of BlipForPretrain. |
Source code in src/unitorch/cli/models/blip/modeling.py
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 |
|
BlipForClassification¤
Tip
core/model/classification/blip
is the section for configuration of BlipForClassification.
Bases: BlipForClassification
BLIP model for classification.
Initialize BlipForClassification.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config_path |
str
|
The path to the model configuration file. |
required |
projection_dim |
int
|
The dimension of the projection head. Defaults to 512. |
512
|
num_classes |
int
|
The number of classes for classification. Defaults to 1. |
1
|
freeze_base_model |
bool
|
Whether to freeze the base model parameters. Defaults to True. |
True
|
gradient_checkpointing |
bool
|
Whether to use gradient checkpointing. Defaults to False. |
False
|
Source code in src/unitorch/cli/models/blip/modeling.py
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 |
|
forward ¤
forward(
input_ids: Tensor,
pixel_values: Tensor,
attention_mask: Tensor = None,
position_ids: Tensor = None,
)
Forward pass of the BlipForClassification model.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
input_ids |
Tensor
|
The input token IDs. |
required |
pixel_values |
Tensor
|
The pixel values of the images. |
required |
attention_mask |
Tensor
|
The attention mask. Defaults to None. |
None
|
position_ids |
Tensor
|
The position IDs. Defaults to None. |
None
|
Returns:
Name | Type | Description |
---|---|---|
ClassificationOutputs |
The classification outputs. |
Source code in src/unitorch/cli/models/blip/modeling.py
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 |
|
from_core_configure
classmethod
¤
from_core_configure(config, **kwargs)
Create an instance of BlipForClassification from a core configuration.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config |
The core configuration. |
required | |
**kwargs |
Additional keyword arguments. |
{}
|
Returns:
Name | Type | Description |
---|---|---|
BlipForClassification |
An instance of BlipForClassification. |
Source code in src/unitorch/cli/models/blip/modeling.py
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 |
|
BlipForTextClassification¤
Tip
core/model/classification/blip/text
is the section for configuration of BlipForTextClassification.
Bases: BlipForTextClassification
BLIP model for text classification.
Initialize BlipForTextClassification.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config_path |
str
|
The path to the model configuration file. |
required |
projection_dim |
int
|
The dimension of the projection head. Defaults to 512. |
512
|
num_classes |
int
|
The number of classes for classification. Defaults to 1. |
1
|
freeze_base_model |
bool
|
Whether to freeze the base model parameters. Defaults to True. |
True
|
gradient_checkpointing |
bool
|
Whether to use gradient checkpointing. Defaults to False. |
False
|
Source code in src/unitorch/cli/models/blip/modeling.py
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 |
|
forward ¤
forward(
input_ids=None,
attention_mask: Optional[Tensor] = None,
position_ids: Optional[Tensor] = None,
)
Forward pass of the BlipForTextClassification model.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
input_ids |
Tensor
|
The input token IDs. Defaults to None. |
None
|
attention_mask |
Tensor
|
The attention mask. Defaults to None. |
None
|
position_ids |
Tensor
|
The position IDs. Defaults to None. |
None
|
Returns:
Name | Type | Description |
---|---|---|
ClassificationOutputs |
The classification outputs. |
Source code in src/unitorch/cli/models/blip/modeling.py
318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 |
|
from_core_configure
classmethod
¤
from_core_configure(config, **kwargs)
Create an instance of BlipForTextClassification from a core configuration.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config |
The core configuration. |
required | |
**kwargs |
Additional keyword arguments. |
{}
|
Returns:
Name | Type | Description |
---|---|---|
BlipForTextClassification |
An instance of BlipForTextClassification. |
Source code in src/unitorch/cli/models/blip/modeling.py
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 |
|
BlipForImageClassification¤
Tip
core/model/classification/blip/image
is the section for configuration of BlipForImageClassification.
Bases: BlipForImageClassification
BLIP model for image classification.
Initialize BlipForImageClassification.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config_path |
str
|
The path to the model configuration file. |
required |
projection_dim |
int
|
The dimension of the projection head. Defaults to 512. |
512
|
num_classes |
int
|
The number of classes for classification. Defaults to 1. |
1
|
freeze_base_model |
bool
|
Whether to freeze the base model parameters. Defaults to True. |
True
|
gradient_checkpointing |
bool
|
Whether to use gradient checkpointing. Defaults to False. |
False
|
Source code in src/unitorch/cli/models/blip/modeling.py
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 |
|
forward ¤
forward(pixel_values: Tensor)
Forward pass of the BlipForImageClassification model.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
pixel_values |
Tensor
|
The pixel values of the images. |
required |
Returns:
Name | Type | Description |
---|---|---|
ClassificationOutputs |
The classification outputs. |
Source code in src/unitorch/cli/models/blip/modeling.py
422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 |
|
from_core_configure
classmethod
¤
from_core_configure(config, **kwargs)
Create an instance of BlipForImageClassification from a core configuration.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config |
The core configuration. |
required | |
**kwargs |
Additional keyword arguments. |
{}
|
Returns:
Name | Type | Description |
---|---|---|
BlipForImageClassification |
An instance of BlipForImageClassification. |
Source code in src/unitorch/cli/models/blip/modeling.py
374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 |
|
BlipForImageCaption¤
Tip
core/model/caption/blip
is the section for configuration of BlipForImageCaption.
Bases: BlipForImageCaption
BLIP model for image captioning.
Initialize BlipForImageCaption.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config_path |
str
|
The path to the model configuration file. |
required |
gradient_checkpointing |
bool
|
Whether to use gradient checkpointing. Defaults to False. |
False
|
Source code in src/unitorch/cli/models/blip/modeling.py
444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 |
|
forward ¤
forward(
pixel_values: Tensor,
input_ids: Tensor,
attention_mask: Optional[Tensor] = None,
)
Forward pass of the BlipForImageCaption model.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
pixel_values |
Tensor
|
The pixel values of the images. |
required |
input_ids |
Tensor
|
The input captions. |
required |
Returns:
Name | Type | Description |
---|---|---|
GenerationOutputs |
The generation outputs. |
Source code in src/unitorch/cli/models/blip/modeling.py
502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 |
|
from_core_configure
classmethod
¤
from_core_configure(config, **kwargs)
Create an instance of BlipForImageCaption from a core configuration.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
config |
The core configuration. |
required | |
**kwargs |
Additional keyword arguments. |
{}
|
Returns:
Name | Type | Description |
---|---|---|
BlipForImageCaption |
An instance of BlipForImageCaption. |
Source code in src/unitorch/cli/models/blip/modeling.py
461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 |
|
generate ¤
generate(
pixel_values: Tensor,
num_beams: Optional[int] = 5,
decoder_start_token_id: Optional[int] = 101,
decoder_end_token_id: Optional[
Union[int, List[int]]
] = 102,
num_return_sequences: Optional[int] = 1,
min_gen_seq_length: Optional[int] = 0,
max_gen_seq_length: Optional[int] = 48,
repetition_penalty: Optional[float] = 1.0,
no_repeat_ngram_size: Optional[int] = 0,
early_stopping: Optional[bool] = True,
length_penalty: Optional[float] = 1.0,
num_beam_groups: Optional[int] = 1,
diversity_penalty: Optional[float] = 0.0,
do_sample: Optional[bool] = False,
temperature: Optional[float] = 1.0,
top_k: Optional[int] = 50,
top_p: Optional[float] = 1.0,
)
Generate captions using the BlipForImageCaption model.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
pixel_values |
Tensor
|
The pixel values of the images. |
required |
num_beams |
int
|
The number of beams for beam search. Defaults to 5. |
5
|
decoder_start_token_id |
int
|
The start token ID for the decoder. Defaults to 30522. |
101
|
decoder_end_token_id |
int or List[int]
|
The end token ID for the decoder. Defaults to 2. |
102
|
num_return_sequences |
int
|
The number of sequences to return. Defaults to 1. |
1
|
min_gen_seq_length |
int
|
The minimum length of generated sequences. Defaults to 0. |
0
|
max_gen_seq_length |
int
|
The maximum length of generated sequences. Defaults to 48. |
48
|
repetition_penalty |
float
|
The repetition penalty. Defaults to 1.0. |
1.0
|
no_repeat_ngram_size |
int
|
The size of n-grams to avoid repeating. Defaults to 0. |
0
|
early_stopping |
bool
|
Whether to perform early stopping. Defaults to True. |
True
|
length_penalty |
float
|
The length penalty. Defaults to 1.0. |
1.0
|
num_beam_groups |
int
|
The number of beam groups. Defaults to 1. |
1
|
diversity_penalty |
float
|
The diversity penalty. Defaults to 0.0. |
0.0
|
do_sample |
bool
|
Whether to use sampling for generation. Defaults to False. |
False
|
temperature |
float
|
The temperature value for sampling. Defaults to 1.0. |
1.0
|
top_k |
int
|
The top-k value for sampling. Defaults to 50. |
50
|
top_p |
float
|
The top-p value for sampling. Defaults to 1.0. |
1.0
|
Returns:
Name | Type | Description |
---|---|---|
GenerationOutputs |
The generation outputs. |
Source code in src/unitorch/cli/models/blip/modeling.py
526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 |
|