You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@lucene.apache.org by "benwtrent (via GitHub)" <gi...@apache.org> on 2024/12/11 16:11:01 UTC

[PR] Adjust test for new random quantization formats [lucene]

benwtrent opened a new pull request, #14058:
URL: https://github.com/apache/lucene/pull/14058

   lucene now randomly adjusts quantization params, etc. and consequently this test sometimes doesn't return exactly a score of `0`. Adjusting to ensure we are never negative.
   
   Related: https://github.com/apache/lucene/pull/14049
   closes: https://github.com/apache/lucene/issues/14051


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org

Re: [PR] Adjust test for new random quantization formats [lucene]

Posted by "msokolov (via GitHub)" <gi...@apache.org>.
msokolov commented on PR #14058:
URL: https://github.com/apache/lucene/pull/14058#issuecomment-2539387026

   oh I see; I think in other cases like this, I disabled the randomness, but easing the constraint seems OK too.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org

Re: [PR] Adjust test for new random quantization formats [lucene]

Posted by "benwtrent (via GitHub)" <gi...@apache.org>.
benwtrent merged PR #14058:
URL: https://github.com/apache/lucene/pull/14058


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org

Re: [PR] Adjust test for new random quantization formats [lucene]

Posted by "msokolov (via GitHub)" <gi...@apache.org>.
msokolov commented on PR #14058:
URL: https://github.com/apache/lucene/pull/14058#issuecomment-2539139518

   I meant previously the test asserted (implicitly) that the value was in [-EPSILON, EPSILON]. You changed it to assert that the value is > 0. I was proposing a more restrictive criterion: that it is in [0, EPSILON].


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org

Re: [PR] Adjust test for new random quantization formats [lucene]

Posted by "benwtrent (via GitHub)" <gi...@apache.org>.
benwtrent commented on PR #14058:
URL: https://github.com/apache/lucene/pull/14058#issuecomment-2539129207

   > do we still also want to assert that it is < EPSILON?
   
   In what way? We care that its greater than `0`. Previously it was already asserting that it was "near" zero via `EPISON`, which indicates `< EPSILON`.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org

Re: [PR] Adjust test for new random quantization formats [lucene]

Posted by "benwtrent (via GitHub)" <gi...@apache.org>.
benwtrent commented on PR #14058:
URL: https://github.com/apache/lucene/pull/14058#issuecomment-2539364258

   @msokolov FYI the test failed because the score was `0.5`


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org