stefanvodita commented on code in PR #14238:
URL: https://github.com/apache/lucene/pull/14238#discussion_r1956984177
##
lucene/CHANGES.txt:
##
@@ -20,6 +20,7 @@ New Features
Improvements
-
+* GITHUB#14238: Improve test coverage of Dynamic Range Faceting.
houserjohn commented on code in PR #14238:
URL: https://github.com/apache/lucene/pull/14238#discussion_r1956936618
##
lucene/facet/src/test/org/apache/lucene/facet/range/TestDynamicRangeUtil.java:
##
@@ -76,13 +77,248 @@ public void
testComputeDynamicNumericRangesWithOneLargeWe
houserjohn commented on code in PR #14238:
URL: https://github.com/apache/lucene/pull/14238#discussion_r1956924426
##
lucene/facet/src/test/org/apache/lucene/facet/range/TestDynamicRangeUtil.java:
##
@@ -76,13 +77,248 @@ public void
testComputeDynamicNumericRangesWithOneLargeWe
houserjohn commented on code in PR #14238:
URL: https://github.com/apache/lucene/pull/14238#discussion_r1956920492
##
lucene/facet/src/test/org/apache/lucene/facet/range/TestDynamicRangeUtil.java:
##
@@ -76,13 +77,248 @@ public void
testComputeDynamicNumericRangesWithOneLargeWe
houserjohn commented on code in PR #14238:
URL: https://github.com/apache/lucene/pull/14238#discussion_r1956920117
##
lucene/facet/src/test/org/apache/lucene/facet/range/TestDynamicRangeUtil.java:
##
@@ -76,13 +77,248 @@ public void
testComputeDynamicNumericRangesWithOneLargeWe
houserjohn commented on code in PR #13914:
URL: https://github.com/apache/lucene/pull/13914#discussion_r1956877159
##
lucene/facet/src/java/org/apache/lucene/facet/range/DynamicRangeUtil.java:
##
@@ -202,66 +208,83 @@ public SegmentOutput(int hitsLength) {
* is used to c
benwtrent closed issue #14206: TestLogMergePolicy#testNoPathologicalMerge
reproducible failure
URL: https://github.com/apache/lucene/issues/14206
--
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
benwtrent closed issue #14206: TestLogMergePolicy#testNoPathologicalMerge
reproducible failure
URL: https://github.com/apache/lucene/issues/14206
--
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
benwtrent merged PR #14245:
URL: https://github.com/apache/lucene/pull/14245
--
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-unsubscr...@lucene.a
msokolov commented on PR #14226:
URL: https://github.com/apache/lucene/pull/14226#issuecomment-2660180526
I don't believe 16 is "special" except in the sense that it happens to be a
sweet spot is this context. We expect that as we increase that per-segment
factor we will get increased reca
benwtrent commented on PR #14226:
URL: https://github.com/apache/lucene/pull/14226#issuecomment-2660166393
> The latency improvement from re-using scores is surprisingly large. I
would have expected costs to be dominated by newly-explored nodes, but this is
cool.
@msokolov this was o
Tim-Brooks commented on PR #14213:
URL: https://github.com/apache/lucene/pull/14213#issuecomment-2660154259
> @Tim-Brooks Could you add an entry in CHANGES.txt? It should be under the
10.2 version, thanks!
I made this change and added some more tests. Let me know if any additional
te
msokolov commented on PR #14226:
URL: https://github.com/apache/lucene/pull/14226#issuecomment-2660150066
The latency improvement from re-using scores is surprisingly large. I would
have expected costs to be dominated by newly-explored nodes, but this is cool.
--
This is an automated mess
gf2121 opened a new pull request, #14244:
URL: https://github.com/apache/lucene/pull/14244
This patch tries to introduce additional relation check to shortcut visiting
within BKD leaves. This helps queries that hit few docs like PointInSetQuery or
narrow RangeQueries on high-cardinality fie
stefanvodita commented on code in PR #14237:
URL: https://github.com/apache/lucene/pull/14237#discussion_r1956459385
##
lucene/demo/src/java/org/apache/lucene/demo/facet/SandboxFacetsExample.java:
##
@@ -130,6 +135,88 @@ void index() throws IOException {
IOUtils.close(index
jpountz commented on PR #13470:
URL: https://github.com/apache/lucene/pull/13470#issuecomment-2659645405
@harenlin I took some freedom to apply my feedback and push it to your
branch. Would you like to take a look and check if it makes sense?
--
This is an automated message from the Apach
benwtrent commented on PR #14160:
URL: https://github.com/apache/lucene/pull/14160#issuecomment-2659586574
Annot & other potpourri I fixed during the prefiltered benchmarking for this
change. https://github.com/mikemccand/luceneutil/pull/337
--
This is an automated message from the Apache
gf2121 commented on PR #14203:
URL: https://github.com/apache/lucene/pull/14203#issuecomment-2659556195
[perf_asm.log](https://github.com/user-attachments/files/18801016/perf_asm.log)
Profile suggests that loops get vectorized.
--
This is an automated message from the Apache Git Se
benwtrent commented on PR #14160:
URL: https://github.com/apache/lucene/pull/14160#issuecomment-2659548003
Nightly has picked up the change:
https://benchmarks.mikemccandless.com/PreFilteredVectorSearch.html
https://github.com/user-attachments/assets/535bbaf7-a38a-4c8e-9cf7-3e7c
gf2121 commented on PR #14203:
URL: https://github.com/apache/lucene/pull/14203#issuecomment-2659531719
Results on my machines are a bit disappointing
```
java -version
openjdk version "23.0.2" 2025-01-21
OpenJDK Runtime Environment (build 23.0.2+7-58)
OpenJDK 64-Bit Server VM
jpountz commented on PR #14203:
URL: https://github.com/apache/lucene/pull/14203#issuecomment-2659510128
Yes, exactly.
--
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 un
gf2121 commented on PR #14203:
URL: https://github.com/apache/lucene/pull/14203#issuecomment-2659505128
Thanks for feedback!
And sorry for my poor english.. Do you mean something like this by `single
batch size of 16 of 32` ?
```
private static void readDelta16(IndexInput in, i
jpountz commented on PR #14163:
URL: https://github.com/apache/lucene/pull/14163#issuecomment-2659489394
This PR will not get merged, but we are interested in fixing the underlying
bug that causes Lucene to not run a merge when it (at least apparently)
obviously should.
--
This is an aut
jpountz commented on PR #14176:
URL: https://github.com/apache/lucene/pull/14176#issuecomment-2659465229
I pushed an annotation.
https://github.com/mikemccand/luceneutil/commit/e07e590ca21b7aacc91f0de9a296f8ffd2042bc3
--
This is an automated message from the Apache Git Service.
To respond
jpountz commented on PR #14203:
URL: https://github.com/apache/lucene/pull/14203#issuecomment-2659455382
Thanks for iterating and running benchmarks. I played with the
micro-benchmark and I get almost the same result if I use a single batch size
of 16 of 32 (AMD Ryzen with AVX2 but no AVX-5
benwtrent closed issue #14233: Hnsw format testRecall failing
URL: https://github.com/apache/lucene/issues/14233
--
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,
benwtrent closed issue #14233: Hnsw format testRecall failing
URL: https://github.com/apache/lucene/issues/14233
--
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,
benwtrent merged PR #14234:
URL: https://github.com/apache/lucene/pull/14234
--
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-unsubscr...@lucene.a
giorgigagnidze16 closed pull request #14242: Simplify operations by avoiding
calculation of extra bitwise operations
URL: https://github.com/apache/lucene/pull/14242
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
giorgigagnidze16 closed issue #14240: Simplify bitwise operations in
DefaultVectorUtilSupport
URL: https://github.com/apache/lucene/issues/14240
--
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 s
rmuir commented on PR #14242:
URL: https://github.com/apache/lucene/pull/14242#issuecomment-2659247757
> Double checking that it works for you too @rmuir?
As long as we fix javadocs of VectorSpecies.loopBound to match :)
Vector operations loops are not the place to optimize or
rmuir commented on PR #14242:
URL: https://github.com/apache/lucene/pull/14242#issuecomment-2659233974
This was written this way (with a bit of redundancy) on purpose to make it
100% clear it is consistent with `VectorSpecies.loopBound`:
> As long as VLENGTH is a power of two, then t
ChrisHegarty commented on PR #14131:
URL: https://github.com/apache/lucene/pull/14131#issuecomment-2659128547
I just committed a rewrite for the cuVS format implementation.
After the rewrite all the BaseKnnVectorsFormatTestCase tests pass. There are
still some lurking intermittent fai
jpountz commented on PR #14242:
URL: https://github.com/apache/lucene/pull/14242#issuecomment-2659075266
Thank you. There shouldn't be any performance difference since I would
expect the compiler to pre-compute the result of `~(4-1)` since these are
constants, so this change is only about r
giorgigagnidze16 opened a new pull request, #14242:
URL: https://github.com/apache/lucene/pull/14242
### Description
A minor change to eliminate unnecessary bitwise calculations, replaced with
actual expected value
Example:
`int upperBound = a.length & ~(4 - 1); ` -> `int upper
giorgigagnidze16 commented on issue #14240:
URL: https://github.com/apache/lucene/issues/14240#issuecomment-2659053672
Figured it out, thanks. Didn't realize people were forking the repos first,
#14242 Here's the PR
--
This is an automated message from the Apache Git Service.
To respond t
jpountz commented on issue #14240:
URL: https://github.com/apache/lucene/issues/14240#issuecomment-2658992237
That shouldn't be required. Have you checked out
https://github.com/apache/lucene/blob/main/CONTRIBUTING.md? Feel free to let me
know if there is a step that isn't clear to you.
-
jpountz opened a new issue, #14241:
URL: https://github.com/apache/lucene/issues/14241
### Description
Searching across multiple fields is a very frequent need, e.g. both Solr and
Elasticsearch allow searching a query string on multiple fields, yet Lucene
doesn't make it easy.
giorgigagnidze16 commented on issue #14240:
URL: https://github.com/apache/lucene/issues/14240#issuecomment-2658975225
@jpountz Sure, but I couldn't figure out how to :) Need to be added to
contributors list somehow?
--
This is an automated message from the Apache Git Service.
To respond
jpountz commented on issue #14240:
URL: https://github.com/apache/lucene/issues/14240#issuecomment-2658971004
Would you like to open a pull request?
--
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 t
stefanvodita commented on code in PR #14238:
URL: https://github.com/apache/lucene/pull/14238#discussion_r1955848183
##
lucene/facet/src/test/org/apache/lucene/facet/range/TestDynamicRangeUtil.java:
##
@@ -76,13 +77,248 @@ public void
testComputeDynamicNumericRangesWithOneLarge
giorgigagnidze16 opened a new issue, #14240:
URL: https://github.com/apache/lucene/issues/14240
### Description
Currently, `DefaultVectorUtilSupport` contains methods that use unnecessary
bitwise operations. These can be replaced with direct values to eliminate extra
computations. Fo
42 matches
Mail list logo
