1. 02 Aug, 2021 1 commit
  2. 28 Jun, 2021 1 commit
  3. 14 Jun, 2021 1 commit
  4. 03 Jun, 2021 1 commit
    • Nicholas Vadivelu's avatar
      Fix weight decay masking in `run_flax_glue.py` (#11964) · 4674061b
      Nicholas Vadivelu authored
      
      
      * Fix weight decay masking in `run_flax_glue.py`
      
      Issues with the previous implementation:
      - The `dict` from `traverse_util.flatten_dict` has keys which are tuples of strings, not one long string with the path separated by periods.
      - `optax.masked` applies the transformation wherever the mask is True, so the masks are flipped.
      - Flax's LayerNorm calls the scale parameter `scale` not `weight`
      
      * Fix formatting with black
      
      * adapt results
      Co-authored-by: default avatarPatrick von Platen <patrick@huggingface.co>
      4674061b
  5. 21 May, 2021 2 commits
  6. 17 May, 2021 1 commit
  7. 14 May, 2021 1 commit
  8. 12 May, 2021 1 commit
  9. 11 May, 2021 1 commit