transformers
54a2361a - Adding _tie_weights() to prediction heads to support low_cpu_mem_usage=True (#29024)

Commit

1 year ago

Adding _tie_weights() to prediction heads to support low_cpu_mem_usage=True (#29024) * Adding _tie_weights() to prediction heads to support low_cpu_mem_usage=True * Testing for the non-safe-tensors case, since the default is safe-tensors already * Running fixup/fix-copies * Adding accelerate annotations to tests

References

#29024 - Adding _tie_weights() to prediction heads to support low_cpu_mem_usage=True

#29969 - [SigLIP] Add fast tokenizer

#32831 - [Docs] Update resources

#33111 - [Backbone] Remove out_features everywhere

#33174 - [Zero-shot image classification pipeline] Remove tokenizer_kwargs

#39821 - Support MetaCLIP 2

#59 - Fix attention mask handling in EoMT-DINOv3 converter

#62 - Add initial DEIMv2 model implementation

#65 - Fix RTDetrV2 sine position embedding ordering

#43710 - [Docs] Add docs for GLM-OCR and fix EomT-DINOv3

Author

hackyon

Parents

ce47582d

transformers 54a2361a - Adding _tie_weights() to prediction heads to support low_cpu_mem_usage=True (#29024)

transformers
54a2361a - Adding _tie_weights() to prediction heads to support low_cpu_mem_usage=True (#29024)