[Maskformer] Add MaskFormerSwin backbone (#20344)

Commit

3 years ago

[Maskformer] Add MaskFormerSwin backbone (#20344) * First draft * Fix backwards compatibility * More fixes * More fixes * Make backbone more general * Improve backbone * Improve test * Fix config checkpoint * Address comments * Use model_type * Address more comments * Fix special model names * Remove MaskFormerSwinModel and MaskFormerSwinPreTrainedModel from main init * Fix typo * Update backbone * Apply suggestion Co-authored-by: Niels Rogge <nielsrogge@Nielss-MacBook-Pro.local>

References

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

#20344 - [Maskformer] Add MaskFormerSwin backbone

#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

NielsRogge

Parents

955780d3

transformers 6dc884ab - [Maskformer] Add MaskFormerSwin backbone (#20344)

transformers
6dc884ab - [Maskformer] Add MaskFormerSwin backbone (#20344)