Repository navigation
Conversation
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds a CPU semantic control for the public EMPF penalty at widths 64, 65 and 2,150. Swapping the two groups changes both conditional spreads while preserving every pooled sample. Joint row/condition permutations and global contraction provide controls. Each call also checks the gradient, application counter and inactive prototype bank.
The swap is detectable at width 64. The two wider sensitivity assertions are strict expected failures for the released observer, so a future fix makes them visible as unexpected passes. The companion Tier 2 benchmark tracks learned quality separately.
Verified locally with the full EMPF test module and the new module: 22 passed, 2 expected failures.
Closes #555. Related to #554.