Add GGUF support to T5-Encoder #36700
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Isotr0py <[email protected]>
Signed-off-by: Isotr0py <[email protected]>
Isotr0py
changed the title
Contributor
|
Thanks @Isotr0py SG 🔥 ! Just have a small question |
MekkCyber
reviewed
Mar 13, 2025
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
SunMarc
approved these changes
Mar 13, 2025
src/transformers/modeling_utils.py
Outdated
Comment on lines
4232
to
4233
| if gguf_file is not None: | ||
| model_kwargs.pop("gguf_file") |
Contributor
Author
There was a problem hiding this comment.
If we use T5EncoderModel to initialize the model class instead of AutoModel:
text_encoder_2 = T5EncoderModel.from_pretrained(
t5_gguf,
gguf_file=t5_file,
torch_dtype=torch.bfloat16,
)Then there will raise an error because gguf_file appeared in model_kwargs:
Traceback (most recent call last):
File "/data/develop-projects/github-repos/transformers/check_t5.py", line 13, in <module>
text_encoder_2 = T5EncoderModel.from_pretrained(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/data/develop-projects/github-repos/transformers/src/transformers/modeling_utils.py", line 272, in _wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/data/develop-projects/github-repos/transformers/src/transformers/modeling_utils.py", line 4377, in from_pretrained
model = cls(config, *model_args, **model_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: T5EncoderModel.__init__() got an unexpected keyword argument 'gguf_file'
I suspect the root issue is in another place, but haven't figured it out yet. 🤔
Member
There was a problem hiding this comment.
hmmm this happens because you passed gguf_file in cls.config_class.from_pretrained( but indeed this is strange since we should pass gguf_file inside that method from the start
Contributor
Author
There was a problem hiding this comment.
hmmm, if we don't pass gguf_file in cls.config_class.from_pretrained(, it will return default T5 config causing parameter shape mismatched:
Traceback (most recent call last):
File "/data/develop-projects/github-repos/transformers/check_t5.py", line 13, in <module>
text_encoder_2 = T5EncoderModel.from_pretrained(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/data/develop-projects/github-repos/transformers/src/transformers/modeling_utils.py", line 272, in _wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/data/develop-projects/github-repos/transformers/src/transformers/modeling_utils.py", line 4430, in from_pretrained
) = cls._load_pretrained_model(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/data/develop-projects/github-repos/transformers/src/transformers/modeling_utils.py", line 4859, in _load_pretrained_model
disk_offload_index, cpu_offload_index = _load_state_dict_into_meta_model(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/data/develop-projects/github-repos/transformers/.venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/data/develop-projects/github-repos/transformers/src/transformers/modeling_utils.py", line 835, in _load_state_dict_into_meta_model
module.load_state_dict(
File "/data/develop-projects/github-repos/transformers/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 2581, in load_state_dict
raise RuntimeError(
RuntimeError: Error(s) in loading state_dict for Linear:
size mismatch for weight: copying a param with shape torch.Size([4096, 4096]) from checkpoint, the shape in current model is torch.Size([512, 512]).
Member
There was a problem hiding this comment.
yeah, this is bit strange that we had to do this only now. Merging for now !
Signed-off-by: Isotr0py <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What does this PR do?
Fixes #33389 (comment)
Before submitting
Pull Request section?
to it if that's the case.
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
@SunMarc @MekkCyber
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.