Avoid float32 overflow when ranking paired MSA rows. - #719
Open
shubhxho wants to merge 1 commit into
Open
Conversation
The rank metric multiplies one MSA row index per chain in float32. With a dozen chains and deep MSAs that product is well past 3.4e38, so every row comes out as inf and argsort no longer orders anything - the pairing keeps an arbitrary 8192 rows instead of the ones highest up the input MSAs. It also produces the "overflow encountered in reduce" warning in issue 236. Sum the logs instead, which gives the same order without the overflow. Signed-off-by: Shubh <shubh@shubhxho.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The rank metric multiplies one MSA row index per chain in float32. With a dozen chains and deep MSAs that product is well past 3.4e38, so every row comes out as inf and argsort no longer orders anything - the pairing keeps an arbitrary 8192 rows instead of the ones highest up the input MSAs. It also produces the "overflow encountered in reduce" warning in issue 236.
Sum the logs instead, which gives the same order without the overflow.