forked from huggingface/diffusers
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathmodeling_utils.py
More file actions
847 lines (729 loc) 路 37.6 KB
/
Copy pathmodeling_utils.py
File metadata and controls
847 lines (729 loc) 路 37.6 KB
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
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
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
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
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
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
# coding=utf-8
# Copyright 2022 The HuggingFace Inc. team.
# Copyright (c) 2022, NVIDIA CORPORATION. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import inspect
import os
from functools import partial
from typing import Callable, List, Optional, Tuple, Union
import torch
from torch import Tensor, device
from huggingface_hub import hf_hub_download
from huggingface_hub.utils import EntryNotFoundError, RepositoryNotFoundError, RevisionNotFoundError
from requests import HTTPError
from .. import __version__
from ..utils import (
CONFIG_NAME,
DIFFUSERS_CACHE,
FLAX_WEIGHTS_NAME,
HF_HUB_OFFLINE,
HUGGINGFACE_CO_RESOLVE_ENDPOINT,
SAFETENSORS_WEIGHTS_NAME,
WEIGHTS_NAME,
is_accelerate_available,
is_safetensors_available,
is_torch_version,
logging,
)
logger = logging.get_logger(__name__)
if is_torch_version(">=", "1.9.0"):
_LOW_CPU_MEM_USAGE_DEFAULT = True
else:
_LOW_CPU_MEM_USAGE_DEFAULT = False
if is_accelerate_available():
import accelerate
from accelerate.utils import set_module_tensor_to_device
from accelerate.utils.versions import is_torch_version
if is_safetensors_available():
import safetensors
def get_parameter_device(parameter: torch.nn.Module):
try:
return next(parameter.parameters()).device
except StopIteration:
# For torch.nn.DataParallel compatibility in PyTorch 1.5
def find_tensor_attributes(module: torch.nn.Module) -> List[Tuple[str, Tensor]]:
tuples = [(k, v) for k, v in module.__dict__.items() if torch.is_tensor(v)]
return tuples
gen = parameter._named_members(get_members_fn=find_tensor_attributes)
first_tuple = next(gen)
return first_tuple[1].device
def get_parameter_dtype(parameter: torch.nn.Module):
try:
return next(parameter.parameters()).dtype
except StopIteration:
# For torch.nn.DataParallel compatibility in PyTorch 1.5
def find_tensor_attributes(module: torch.nn.Module) -> List[Tuple[str, Tensor]]:
tuples = [(k, v) for k, v in module.__dict__.items() if torch.is_tensor(v)]
return tuples
gen = parameter._named_members(get_members_fn=find_tensor_attributes)
first_tuple = next(gen)
return first_tuple[1].dtype
def load_state_dict(checkpoint_file: Union[str, os.PathLike]):
"""
Reads a checkpoint file, returning properly formatted errors if they arise.
"""
try:
if os.path.basename(checkpoint_file) == WEIGHTS_NAME:
return torch.load(checkpoint_file, map_location="cpu")
else:
return safetensors.torch.load_file(checkpoint_file, device="cpu")
except Exception as e:
try:
with open(checkpoint_file) as f:
if f.read().startswith("version"):
raise OSError(
"You seem to have cloned a repository without having git-lfs installed. Please install "
"git-lfs and run `git lfs install` followed by `git lfs pull` in the folder "
"you cloned."
)
else:
raise ValueError(
f"Unable to locate the file {checkpoint_file} which is necessary to load this pretrained "
"model. Make sure you have saved the model properly."
) from e
except (UnicodeDecodeError, ValueError):
raise OSError(
f"Unable to load weights from checkpoint file for '{checkpoint_file}' "
f"at '{checkpoint_file}'. "
"If you tried to load a PyTorch model from a TF 2.0 checkpoint, please set from_tf=True."
)
def _load_state_dict_into_model(model_to_load, state_dict):
# Convert old format to new format if needed from a PyTorch state_dict
# copy state_dict so _load_from_state_dict can modify it
state_dict = state_dict.copy()
error_msgs = []
# PyTorch's `_load_from_state_dict` does not copy parameters in a module's descendants
# so we need to apply the function recursively.
def load(module: torch.nn.Module, prefix=""):
args = (state_dict, prefix, {}, True, [], [], error_msgs)
module._load_from_state_dict(*args)
for name, child in module._modules.items():
if child is not None:
load(child, prefix + name + ".")
load(model_to_load)
return error_msgs
class ModelMixin(torch.nn.Module):
r"""
Base class for all models.
[`ModelMixin`] takes care of storing the configuration of the models and handles methods for loading, downloading
and saving models.
- **config_name** ([`str`]) -- A filename under which the model should be stored when calling
[`~models.ModelMixin.save_pretrained`].
"""
config_name = CONFIG_NAME
_automatically_saved_args = ["_diffusers_version", "_class_name", "_name_or_path"]
_supports_gradient_checkpointing = False
def __init__(self):
super().__init__()
@property
def is_gradient_checkpointing(self) -> bool:
"""
Whether gradient checkpointing is activated for this model or not.
Note that in other frameworks this feature can be referred to as "activation checkpointing" or "checkpoint
activations".
"""
return any(hasattr(m, "gradient_checkpointing") and m.gradient_checkpointing for m in self.modules())
def enable_gradient_checkpointing(self):
"""
Activates gradient checkpointing for the current model.
Note that in other frameworks this feature can be referred to as "activation checkpointing" or "checkpoint
activations".
"""
if not self._supports_gradient_checkpointing:
raise ValueError(f"{self.__class__.__name__} does not support gradient checkpointing.")
self.apply(partial(self._set_gradient_checkpointing, value=True))
def disable_gradient_checkpointing(self):
"""
Deactivates gradient checkpointing for the current model.
Note that in other frameworks this feature can be referred to as "activation checkpointing" or "checkpoint
activations".
"""
if self._supports_gradient_checkpointing:
self.apply(partial(self._set_gradient_checkpointing, value=False))
def set_use_memory_efficient_attention_xformers(
self, valid: bool, attention_op: Optional[Callable] = None
) -> None:
# Recursively walk through all the children.
# Any children which exposes the set_use_memory_efficient_attention_xformers method
# gets the message
def fn_recursive_set_mem_eff(module: torch.nn.Module):
if hasattr(module, "set_use_memory_efficient_attention_xformers"):
module.set_use_memory_efficient_attention_xformers(valid, attention_op)
for child in module.children():
fn_recursive_set_mem_eff(child)
for module in self.children():
if isinstance(module, torch.nn.Module):
fn_recursive_set_mem_eff(module)
def enable_xformers_memory_efficient_attention(self, attention_op: Optional[Callable] = None):
r"""
Enable memory efficient attention as implemented in xformers.
When this option is enabled, you should observe lower GPU memory usage and a potential speed up at inference
time. Speed up at training time is not guaranteed.
Warning: When Memory Efficient Attention and Sliced attention are both enabled, the Memory Efficient Attention
is used.
Parameters:
attention_op (`Callable`, *optional*):
Override the default `None` operator for use as `op` argument to the
[`memory_efficient_attention()`](https://facebookresearch.github.io/xformers/components/ops.html#xformers.ops.memory_efficient_attention)
function of xFormers.
Examples:
```py
>>> import torch
>>> from diffusers import UNet2DConditionModel
>>> from xformers.ops import MemoryEfficientAttentionFlashAttentionOp
>>> model = UNet2DConditionModel.from_pretrained(
... "stabilityai/stable-diffusion-2-1", subfolder="unet", torch_dtype=torch.float16
... )
>>> model = model.to("cuda")
>>> model.enable_xformers_memory_efficient_attention(attention_op=MemoryEfficientAttentionFlashAttentionOp)
```
"""
self.set_use_memory_efficient_attention_xformers(True, attention_op)
def disable_xformers_memory_efficient_attention(self):
r"""
Disable memory efficient attention as implemented in xformers.
"""
self.set_use_memory_efficient_attention_xformers(False)
def save_pretrained(
self,
save_directory: Union[str, os.PathLike],
is_main_process: bool = True,
save_function: Callable = None,
safe_serialization: bool = False,
):
"""
Save a model and its configuration file to a directory, so that it can be re-loaded using the
`[`~models.ModelMixin.from_pretrained`]` class method.
Arguments:
save_directory (`str` or `os.PathLike`):
Directory to which to save. Will be created if it doesn't exist.
is_main_process (`bool`, *optional*, defaults to `True`):
Whether the process calling this is the main process or not. Useful when in distributed training like
TPUs and need to call this function on all processes. In this case, set `is_main_process=True` only on
the main process to avoid race conditions.
save_function (`Callable`):
The function to use to save the state dictionary. Useful on distributed training like TPUs when one
need to replace `torch.save` by another method. Can be configured with the environment variable
`DIFFUSERS_SAVE_MODE`.
safe_serialization (`bool`, *optional*, defaults to `False`):
Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
"""
if safe_serialization and not is_safetensors_available():
raise ImportError("`safe_serialization` requires the `safetensors library: `pip install safetensors`.")
if os.path.isfile(save_directory):
logger.error(f"Provided path ({save_directory}) should be a directory, not a file")
return
if save_function is None:
save_function = safetensors.torch.save_file if safe_serialization else torch.save
os.makedirs(save_directory, exist_ok=True)
model_to_save = self
# Attach architecture to the config
# Save the config
if is_main_process:
model_to_save.save_config(save_directory)
# Save the model
state_dict = model_to_save.state_dict()
weights_name = SAFETENSORS_WEIGHTS_NAME if safe_serialization else WEIGHTS_NAME
# Save the model
save_function(state_dict, os.path.join(save_directory, weights_name))
logger.info(f"Model weights saved in {os.path.join(save_directory, weights_name)}")
@classmethod
def from_pretrained(cls, pretrained_model_name_or_path: Optional[Union[str, os.PathLike]], **kwargs):
r"""
Instantiate a pretrained pytorch model from a pre-trained model configuration.
The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated). To train
the model, you should first set it back in training mode with `model.train()`.
The warning *Weights from XXX not initialized from pretrained model* means that the weights of XXX do not come
pretrained with the rest of the model. It is up to you to train those weights with a downstream fine-tuning
task.
The warning *Weights from XXX not used in YYY* means that the layer XXX is not used by YYY, therefore those
weights are discarded.
Parameters:
pretrained_model_name_or_path (`str` or `os.PathLike`, *optional*):
Can be either:
- A string, the *model id* of a pretrained model hosted inside a model repo on huggingface.co.
Valid model ids should have an organization name, like `google/ddpm-celebahq-256`.
- A path to a *directory* containing model weights saved using [`~ModelMixin.save_config`], e.g.,
`./my_model_directory/`.
cache_dir (`Union[str, os.PathLike]`, *optional*):
Path to a directory in which a downloaded pretrained model configuration should be cached if the
standard cache should not be used.
torch_dtype (`str` or `torch.dtype`, *optional*):
Override the default `torch.dtype` and load the model under this dtype. If `"auto"` is passed the dtype
will be automatically derived from the model's weights.
force_download (`bool`, *optional*, defaults to `False`):
Whether or not to force the (re-)download of the model weights and configuration files, overriding the
cached versions if they exist.
resume_download (`bool`, *optional*, defaults to `False`):
Whether or not to delete incompletely received files. Will attempt to resume the download if such a
file exists.
proxies (`Dict[str, str]`, *optional*):
A dictionary of proxy servers to use by protocol or endpoint, e.g., `{'http': 'foo.bar:3128',
'http://hostname': 'foo.bar:4012'}`. The proxies are used on each request.
output_loading_info(`bool`, *optional*, defaults to `False`):
Whether or not to also return a dictionary containing missing keys, unexpected keys and error messages.
local_files_only(`bool`, *optional*, defaults to `False`):
Whether or not to only look at local files (i.e., do not try to download the model).
use_auth_token (`str` or *bool*, *optional*):
The token to use as HTTP bearer authorization for remote files. If `True`, will use the token generated
when running `diffusers-cli login` (stored in `~/.huggingface`).
revision (`str`, *optional*, defaults to `"main"`):
The specific model version to use. It can be a branch name, a tag name, or a commit id, since we use a
git-based system for storing models and other artifacts on huggingface.co, so `revision` can be any
identifier allowed by git.
from_flax (`bool`, *optional*, defaults to `False`):
Load the model weights from a Flax checkpoint save file.
subfolder (`str`, *optional*, defaults to `""`):
In case the relevant files are located inside a subfolder of the model repo (either remote in
huggingface.co or downloaded locally), you can specify the folder name here.
mirror (`str`, *optional*):
Mirror source to accelerate downloads in China. If you are from China and have an accessibility
problem, you can set this option to resolve it. Note that we do not guarantee the timeliness or safety.
Please refer to the mirror site for more information.
device_map (`str` or `Dict[str, Union[int, str, torch.device]]`, *optional*):
A map that specifies where each submodule should go. It doesn't need to be refined to each
parameter/buffer name, once a given module name is inside, every submodule of it will be sent to the
same device.
To have Accelerate compute the most optimized `device_map` automatically, set `device_map="auto"`. For
more information about each option see [designing a device
map](https://hf.co/docs/accelerate/main/en/usage_guides/big_modeling#designing-a-device-map).
low_cpu_mem_usage (`bool`, *optional*, defaults to `True` if torch version >= 1.9.0 else `False`):
Speed up model loading by not initializing the weights and only loading the pre-trained weights. This
also tries to not use more than 1x model size in CPU memory (including peak memory) while loading the
model. This is only supported when torch version >= 1.9.0. If you are using an older version of torch,
setting this argument to `True` will raise an error.
<Tip>
It is required to be logged in (`huggingface-cli login`) when you want to use private or [gated
models](https://huggingface.co/docs/hub/models-gated#gated-models).
</Tip>
<Tip>
Activate the special ["offline-mode"](https://huggingface.co/diffusers/installation.html#offline-mode) to use
this method in a firewalled environment.
</Tip>
"""
cache_dir = kwargs.pop("cache_dir", DIFFUSERS_CACHE)
ignore_mismatched_sizes = kwargs.pop("ignore_mismatched_sizes", False)
force_download = kwargs.pop("force_download", False)
from_flax = kwargs.pop("from_flax", False)
resume_download = kwargs.pop("resume_download", False)
proxies = kwargs.pop("proxies", None)
output_loading_info = kwargs.pop("output_loading_info", False)
local_files_only = kwargs.pop("local_files_only", HF_HUB_OFFLINE)
use_auth_token = kwargs.pop("use_auth_token", None)
revision = kwargs.pop("revision", None)
torch_dtype = kwargs.pop("torch_dtype", None)
subfolder = kwargs.pop("subfolder", None)
device_map = kwargs.pop("device_map", None)
low_cpu_mem_usage = kwargs.pop("low_cpu_mem_usage", _LOW_CPU_MEM_USAGE_DEFAULT)
if low_cpu_mem_usage and not is_accelerate_available():
low_cpu_mem_usage = False
logger.warning(
"Cannot initialize model with low cpu memory usage because `accelerate` was not found in the"
" environment. Defaulting to `low_cpu_mem_usage=False`. It is strongly recommended to install"
" `accelerate` for faster and less memory-intense model loading. You can do so with: \n```\npip"
" install accelerate\n```\n."
)
if device_map is not None and not is_accelerate_available():
raise NotImplementedError(
"Loading and dispatching requires `accelerate`. Please make sure to install accelerate or set"
" `device_map=None`. You can install accelerate with `pip install accelerate`."
)
# Check if we can handle device_map and dispatching the weights
if device_map is not None and not is_torch_version(">=", "1.9.0"):
raise NotImplementedError(
"Loading and dispatching requires torch >= 1.9.0. Please either update your PyTorch version or set"
" `device_map=None`."
)
if low_cpu_mem_usage is True and not is_torch_version(">=", "1.9.0"):
raise NotImplementedError(
"Low memory initialization requires torch >= 1.9.0. Please either update your PyTorch version or set"
" `low_cpu_mem_usage=False`."
)
if low_cpu_mem_usage is False and device_map is not None:
raise ValueError(
f"You cannot set `low_cpu_mem_usage` to `False` while using device_map={device_map} for loading and"
" dispatching. Please make sure to set `low_cpu_mem_usage=True`."
)
user_agent = {
"diffusers": __version__,
"file_type": "model",
"framework": "pytorch",
}
# Load config if we don't provide a configuration
config_path = pretrained_model_name_or_path
# This variable will flag if we're loading a sharded checkpoint. In this case the archive file is just the
# Load model
model_file = None
if from_flax:
model_file = _get_model_file(
pretrained_model_name_or_path,
weights_name=FLAX_WEIGHTS_NAME,
cache_dir=cache_dir,
force_download=force_download,
resume_download=resume_download,
proxies=proxies,
local_files_only=local_files_only,
use_auth_token=use_auth_token,
revision=revision,
subfolder=subfolder,
user_agent=user_agent,
)
config, unused_kwargs = cls.load_config(
config_path,
cache_dir=cache_dir,
return_unused_kwargs=True,
force_download=force_download,
resume_download=resume_download,
proxies=proxies,
local_files_only=local_files_only,
use_auth_token=use_auth_token,
revision=revision,
subfolder=subfolder,
device_map=device_map,
**kwargs,
)
model = cls.from_config(config, **unused_kwargs)
# Convert the weights
from .modeling_pytorch_flax_utils import load_flax_checkpoint_in_pytorch_model
model = load_flax_checkpoint_in_pytorch_model(model, model_file)
else:
if is_safetensors_available():
try:
model_file = _get_model_file(
pretrained_model_name_or_path,
weights_name=SAFETENSORS_WEIGHTS_NAME,
cache_dir=cache_dir,
force_download=force_download,
resume_download=resume_download,
proxies=proxies,
local_files_only=local_files_only,
use_auth_token=use_auth_token,
revision=revision,
subfolder=subfolder,
user_agent=user_agent,
)
except:
pass
if model_file is None:
model_file = _get_model_file(
pretrained_model_name_or_path,
weights_name=WEIGHTS_NAME,
cache_dir=cache_dir,
force_download=force_download,
resume_download=resume_download,
proxies=proxies,
local_files_only=local_files_only,
use_auth_token=use_auth_token,
revision=revision,
subfolder=subfolder,
user_agent=user_agent,
)
if low_cpu_mem_usage:
# Instantiate model with empty weights
with accelerate.init_empty_weights():
config, unused_kwargs = cls.load_config(
config_path,
cache_dir=cache_dir,
return_unused_kwargs=True,
force_download=force_download,
resume_download=resume_download,
proxies=proxies,
local_files_only=local_files_only,
use_auth_token=use_auth_token,
revision=revision,
subfolder=subfolder,
device_map=device_map,
**kwargs,
)
model = cls.from_config(config, **unused_kwargs)
# if device_map is None, load the state dict and move the params from meta device to the cpu
if device_map is None:
param_device = "cpu"
state_dict = load_state_dict(model_file)
# move the params from meta device to cpu
for param_name, param in state_dict.items():
accepts_dtype = "dtype" in set(
inspect.signature(set_module_tensor_to_device).parameters.keys()
)
if accepts_dtype:
set_module_tensor_to_device(
model, param_name, param_device, value=param, dtype=torch_dtype
)
else:
set_module_tensor_to_device(model, param_name, param_device, value=param)
else: # else let accelerate handle loading and dispatching.
# Load weights and dispatch according to the device_map
# by deafult the device_map is None and the weights are loaded on the CPU
accelerate.load_checkpoint_and_dispatch(model, model_file, device_map, dtype=torch_dtype)
loading_info = {
"missing_keys": [],
"unexpected_keys": [],
"mismatched_keys": [],
"error_msgs": [],
}
else:
config, unused_kwargs = cls.load_config(
config_path,
cache_dir=cache_dir,
return_unused_kwargs=True,
force_download=force_download,
resume_download=resume_download,
proxies=proxies,
local_files_only=local_files_only,
use_auth_token=use_auth_token,
revision=revision,
subfolder=subfolder,
device_map=device_map,
**kwargs,
)
model = cls.from_config(config, **unused_kwargs)
state_dict = load_state_dict(model_file)
model, missing_keys, unexpected_keys, mismatched_keys, error_msgs = cls._load_pretrained_model(
model,
state_dict,
model_file,
pretrained_model_name_or_path,
ignore_mismatched_sizes=ignore_mismatched_sizes,
)
loading_info = {
"missing_keys": missing_keys,
"unexpected_keys": unexpected_keys,
"mismatched_keys": mismatched_keys,
"error_msgs": error_msgs,
}
if torch_dtype is not None and not isinstance(torch_dtype, torch.dtype):
raise ValueError(
f"{torch_dtype} needs to be of type `torch.dtype`, e.g. `torch.float16`, but is {type(torch_dtype)}."
)
elif torch_dtype is not None:
model = model.to(torch_dtype)
model.register_to_config(_name_or_path=pretrained_model_name_or_path)
# Set model in evaluation mode to deactivate DropOut modules by default
model.eval()
if output_loading_info:
return model, loading_info
return model
@classmethod
def _load_pretrained_model(
cls,
model,
state_dict,
resolved_archive_file,
pretrained_model_name_or_path,
ignore_mismatched_sizes=False,
):
# Retrieve missing & unexpected_keys
model_state_dict = model.state_dict()
loaded_keys = [k for k in state_dict.keys()]
expected_keys = list(model_state_dict.keys())
original_loaded_keys = loaded_keys
missing_keys = list(set(expected_keys) - set(loaded_keys))
unexpected_keys = list(set(loaded_keys) - set(expected_keys))
# Make sure we are able to load base models as well as derived models (with heads)
model_to_load = model
def _find_mismatched_keys(
state_dict,
model_state_dict,
loaded_keys,
ignore_mismatched_sizes,
):
mismatched_keys = []
if ignore_mismatched_sizes:
for checkpoint_key in loaded_keys:
model_key = checkpoint_key
if (
model_key in model_state_dict
and state_dict[checkpoint_key].shape != model_state_dict[model_key].shape
):
mismatched_keys.append(
(checkpoint_key, state_dict[checkpoint_key].shape, model_state_dict[model_key].shape)
)
del state_dict[checkpoint_key]
return mismatched_keys
if state_dict is not None:
# Whole checkpoint
mismatched_keys = _find_mismatched_keys(
state_dict,
model_state_dict,
original_loaded_keys,
ignore_mismatched_sizes,
)
error_msgs = _load_state_dict_into_model(model_to_load, state_dict)
if len(error_msgs) > 0:
error_msg = "\n\t".join(error_msgs)
if "size mismatch" in error_msg:
error_msg += (
"\n\tYou may consider adding `ignore_mismatched_sizes=True` in the model `from_pretrained` method."
)
raise RuntimeError(f"Error(s) in loading state_dict for {model.__class__.__name__}:\n\t{error_msg}")
if len(unexpected_keys) > 0:
logger.warning(
f"Some weights of the model checkpoint at {pretrained_model_name_or_path} were not used when"
f" initializing {model.__class__.__name__}: {unexpected_keys}\n- This IS expected if you are"
f" initializing {model.__class__.__name__} from the checkpoint of a model trained on another task"
" or with another architecture (e.g. initializing a BertForSequenceClassification model from a"
" BertForPreTraining model).\n- This IS NOT expected if you are initializing"
f" {model.__class__.__name__} from the checkpoint of a model that you expect to be exactly"
" identical (initializing a BertForSequenceClassification model from a"
" BertForSequenceClassification model)."
)
else:
logger.info(f"All model checkpoint weights were used when initializing {model.__class__.__name__}.\n")
if len(missing_keys) > 0:
logger.warning(
f"Some weights of {model.__class__.__name__} were not initialized from the model checkpoint at"
f" {pretrained_model_name_or_path} and are newly initialized: {missing_keys}\nYou should probably"
" TRAIN this model on a down-stream task to be able to use it for predictions and inference."
)
elif len(mismatched_keys) == 0:
logger.info(
f"All the weights of {model.__class__.__name__} were initialized from the model checkpoint at"
f" {pretrained_model_name_or_path}.\nIf your task is similar to the task the model of the"
f" checkpoint was trained on, you can already use {model.__class__.__name__} for predictions"
" without further training."
)
if len(mismatched_keys) > 0:
mismatched_warning = "\n".join(
[
f"- {key}: found shape {shape1} in the checkpoint and {shape2} in the model instantiated"
for key, shape1, shape2 in mismatched_keys
]
)
logger.warning(
f"Some weights of {model.__class__.__name__} were not initialized from the model checkpoint at"
f" {pretrained_model_name_or_path} and are newly initialized because the shapes did not"
f" match:\n{mismatched_warning}\nYou should probably TRAIN this model on a down-stream task to be"
" able to use it for predictions and inference."
)
return model, missing_keys, unexpected_keys, mismatched_keys, error_msgs
@property
def device(self) -> device:
"""
`torch.device`: The device on which the module is (assuming that all the module parameters are on the same
device).
"""
return get_parameter_device(self)
@property
def dtype(self) -> torch.dtype:
"""
`torch.dtype`: The dtype of the module (assuming that all the module parameters have the same dtype).
"""
return get_parameter_dtype(self)
def num_parameters(self, only_trainable: bool = False, exclude_embeddings: bool = False) -> int:
"""
Get number of (optionally, trainable or non-embeddings) parameters in the module.
Args:
only_trainable (`bool`, *optional*, defaults to `False`):
Whether or not to return only the number of trainable parameters
exclude_embeddings (`bool`, *optional*, defaults to `False`):
Whether or not to return only the number of non-embeddings parameters
Returns:
`int`: The number of parameters.
"""
if exclude_embeddings:
embedding_param_names = [
f"{name}.weight"
for name, module_type in self.named_modules()
if isinstance(module_type, torch.nn.Embedding)
]
non_embedding_parameters = [
parameter for name, parameter in self.named_parameters() if name not in embedding_param_names
]
return sum(p.numel() for p in non_embedding_parameters if p.requires_grad or not only_trainable)
else:
return sum(p.numel() for p in self.parameters() if p.requires_grad or not only_trainable)
def _get_model_file(
pretrained_model_name_or_path,
*,
weights_name,
subfolder,
cache_dir,
force_download,
proxies,
resume_download,
local_files_only,
use_auth_token,
user_agent,
revision,
):
pretrained_model_name_or_path = str(pretrained_model_name_or_path)
if os.path.isfile(pretrained_model_name_or_path):
return pretrained_model_name_or_path
elif os.path.isdir(pretrained_model_name_or_path):
if os.path.isfile(os.path.join(pretrained_model_name_or_path, weights_name)):
# Load from a PyTorch checkpoint
model_file = os.path.join(pretrained_model_name_or_path, weights_name)
return model_file
elif subfolder is not None and os.path.isfile(
os.path.join(pretrained_model_name_or_path, subfolder, weights_name)
):
model_file = os.path.join(pretrained_model_name_or_path, subfolder, weights_name)
return model_file
else:
raise EnvironmentError(
f"Error no file named {weights_name} found in directory {pretrained_model_name_or_path}."
)
else:
try:
# Load from URL or cache if already cached
model_file = hf_hub_download(
pretrained_model_name_or_path,
filename=weights_name,
cache_dir=cache_dir,
force_download=force_download,
proxies=proxies,
resume_download=resume_download,
local_files_only=local_files_only,
use_auth_token=use_auth_token,
user_agent=user_agent,
subfolder=subfolder,
revision=revision,
)
return model_file
except RepositoryNotFoundError:
raise EnvironmentError(
f"{pretrained_model_name_or_path} is not a local folder and is not a valid model identifier "
"listed on 'https://huggingface.co/models'\nIf this is a private repository, make sure to pass a "
"token having permission to this repo with `use_auth_token` or log in with `huggingface-cli "
"login`."
)
except RevisionNotFoundError:
raise EnvironmentError(
f"{revision} is not a valid git identifier (branch name, tag name or commit id) that exists for "
"this model name. Check the model page at "
f"'https://huggingface.co/{pretrained_model_name_or_path}' for available revisions."
)
except EntryNotFoundError:
raise EnvironmentError(
f"{pretrained_model_name_or_path} does not appear to have a file named {weights_name}."
)
except HTTPError as err:
raise EnvironmentError(
f"There was a specific connection error when trying to load {pretrained_model_name_or_path}:\n{err}"
)
except ValueError:
raise EnvironmentError(
f"We couldn't connect to '{HUGGINGFACE_CO_RESOLVE_ENDPOINT}' to load this model, couldn't find it"
f" in the cached files and it looks like {pretrained_model_name_or_path} is not the path to a"
f" directory containing a file named {weights_name} or"
" \nCheckout your internet connection or see how to run the library in"
" offline mode at 'https://huggingface.co/docs/diffusers/installation#offline-mode'."
)
except EnvironmentError:
raise EnvironmentError(
f"Can't load the model for '{pretrained_model_name_or_path}'. If you were trying to load it from "
"'https://huggingface.co/models', make sure you don't have a local directory with the same name. "
f"Otherwise, make sure '{pretrained_model_name_or_path}' is the correct path to a directory "
f"containing a file named {weights_name}"
)