[bitsandbbytes] follow-ups #9730
Merged
+123
−65
Conversation
This file contains 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
|
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. |
DN6
reviewed
Oct 21, 2024
src/diffusers/quantizers/base.py
Outdated
| """ | ||
| checks if the quantized param has expected shape. | ||
| """ | ||
| if not hasattr(self, "check_quantized_param_shape"): |
There was a problem hiding this comment.
Think this should be?
Suggested change
| if not hasattr(self, "check_quantized_param_shape"): | |
| if not hasattr(self, "check_if_quantized_param"):: |
There was a problem hiding this comment.
The method is checking for itself? That can't be the case right? Because the subclass inherits it? So it should already be in there?
There was a problem hiding this comment.
My bad. Have updated in 3298a04.
sayakpaul
commented
Oct 22, 2024
| Once a model is quantized, you can push the model to the Hub with the [`~ModelMixin.push_to_hub`] method. The quantization `config.json` file is pushed first, followed by the quantized model weights. | ||
|
|
||
| ```py | ||
| from diffusers import FluxTransformer2DModel, BitsAndBytesConfig | ||
|
|
||
| quantization_config = BitsAndBytesConfig(load_in_8bit=True) | ||
|
|
||
| model_8bit = FluxTransformer2DModel.from_pretrained( | ||
| "black-forest-labs/FLUX.1-dev", | ||
| subfolder="transformer", | ||
| quantization_config=quantization_config | ||
| ) | ||
| ``` | ||
| Once a model is quantized, you can push the model to the Hub with the [`~ModelMixin.push_to_hub`] method. The quantization `config.json` file is pushed first, followed by the quantized model weights. You can also save the serialized 4-bit models locally with [`~ModelMixin.save_pretrained`]. |
There was a problem hiding this comment.
To unify the content between 8bit and 4bit hfoption.
sayakpaul
commented
Oct 22, 2024
| and hf_quantizer.check_if_quantized_param(model, param, param_name, state_dict, param_device=device) | ||
| ): | ||
| hf_quantizer.check_quantized_param_shape(param_name, empty_state_dict[param_name].shape, param.shape) | ||
| elif not is_quant_method_bnb: |
There was a problem hiding this comment.
Could have done with else but I think it's a tad bit safer.
DN6
approved these changes
Oct 22, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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?
bitsandbytes#9213 (comment)hf-internal-testing