transformers
c95f8470 - Clip floating point constants to bf16 range to avoid inf conversion (#20605)

Commit

3 years ago

Clip floating point constants to bf16 range to avoid inf conversion (#20605) Co-authored-by: EC2 Default User <ec2-user@ip-172-31-40-169.us-west-2.compute.internal>

References

#19449 - [WIP] Fix weights initialization of several vision models

#20605 - Clip floating point constants to bf16 range to avoid inf conversion

#27720 - Add common processor tests

#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

Author

sangeethabal

Parents

f68796bd

transformers c95f8470 - Clip floating point constants to bf16 range to avoid inf conversion (#20605)

transformers
c95f8470 - Clip floating point constants to bf16 range to avoid inf conversion (#20605)