Document None defaults for MultiHeadAttention args#5470
Open
adityasingh2400 wants to merge 1 commit into
Open
Conversation
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
Several MultiHeadAttention args accept None, but the docstring did not say what happens in that case. Spell it out for qkv_features, out_features (both default to inputs_q.shape[-1]), deterministic (falls back to the __call__ kwarg of the same name), and precision (uses lax.Precision.DEFAULT). Apply the same expansion to the NNX twin in flax/nnx/nn/attention.py. Fixes google#4182
adityasingh2400
force-pushed
the
doc-multihead-attention-none-defaults-4182
branch
from
9cd83f8 to
0581162
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Several
MultiHeadAttentionattributes acceptNone, but the docstring did not say what happens when you pass it. This fills in the gaps for the four args called out in #4182:qkv_features: defaults toinputs_q.shape[-1](linen) /in_features(nnx).out_features: defaults toinputs_q.shape[-1](linen) /in_features(nnx).deterministic: falls back to thedeterministicargument passed to__call__.precision: usesjax.lax.Precision.DEFAULT.Applied symmetrically to
flax/linen/attention.py(MultiHeadDotProductAttentionand itsMultiHeadAttentionalias) and the NNX twin inflax/nnx/nn/attention.py.Pure docstring change, no behavior change.
Fixes #4182
Test plan
python3 -c 'import ast; ast.parse(...)'on both files (parses clean).