Layer2NoNorm uses same mix and delta value during eval mode as during training (useful for debugging or analysis of Layer2NoNorm transition) - Githubissues

huggingface / nn_pruning

Prune a model while finetuning or training.

Apache License 2.0

394 stars 58 forks source link

Layer2NoNorm uses same mix and delta value during eval mode as during training (useful for debugging or analysis of Layer2NoNorm transition) #13

Closed echarlaix closed 3 years ago

madlag commented 3 years ago

Perfect !