transformers
b1ec7454 - Fix RT-DETR inference with float16 and bfloat16 (#31639)

Commit

1 year ago

Fix RT-DETR inference with float16 and bfloat16 (#31639) * [run_slow] rt_detr * Fix positional embeddings and anchors dtypes * [run slow] rt_detr * Apply suggestions from code review Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com> * Fixup --------- Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

References

#62 - Add initial DEIMv2 model implementation

#58 - Add EoMT DINOv3 model

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

#32831 - [Docs] Update resources

#29969 - [SigLIP] Add fast tokenizer

#41212 - Add EoMT with DINOv3 backbone

#39821 - Support MetaCLIP 2

#33111 - [Backbone] Remove out_features everywhere

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

#31639 - Fix RT-DETR inference with float16 and bfloat16

Author

qubvel

Parents

3f93fd06

transformers b1ec7454 - Fix RT-DETR inference with float16 and bfloat16 (#31639)

transformers
b1ec7454 - Fix RT-DETR inference with float16 and bfloat16 (#31639)