transformers
2f59ad16 - [trainer/deepspeed] load_best_model (reimplement re-init) (#17151)

Commit

3 years ago

[trainer/deepspeed] load_best_model (reimplement re-init) (#17151) * [trainer/deepspeed] load_best_model * to sync with DS PR #1947 * simplify * rework load_best_model test * cleanup * bump deepspeed>=0.6.5 Co-authored-by: Olatunji Ruwase <olruwase@microsoft.com>

References

#17151 - [trainer/deepspeed] load_best_model (reimplement re-init)

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

#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

#41212 - Add EoMT with DINOv3 backbone

#62 - Add initial DEIMv2 model implementation

Author

stas00

Parents

046c5ea9

transformers 2f59ad16 - [trainer/deepspeed] load_best_model (reimplement re-init) (#17151)

transformers
2f59ad16 - [trainer/deepspeed] load_best_model (reimplement re-init) (#17151)