Closed gbg141 closed 3 months ago
Check out this pull request on
See visual diffs & provide feedback on Jupyter Notebooks.
Powered by ReviewNB
All modified and coverable lines are covered by tests :white_check_mark:
Project coverage is 97.37%. Comparing base (
2267768
) to head (3652f91
).
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
Standardizing the CAN attention mechanism. Now, as in other attention-based models, the argument
out_channels
actually represents the final number of output channels --i.e., it must be divisible byheads
to successfully perform the attention mechanism.Previously the actual output channels were computed as
heads * out_channels
, which could be misleading.