dweiss commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759046099
So it seems something has changed and the javaCompiler property is now
always non-null, causing -X options to be passed to the forked compiler
(instead of -J...). I've changed this c
MarcusSorealheis commented on issue #11507:
URL: https://github.com/apache/lucene/issues/11507#issuecomment-1759050979
I think we should close it for sure.
--
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
vsop-479 commented on PR #11888:
URL: https://github.com/apache/lucene/pull/11888#issuecomment-1759050886
Append some performance data. Note that the results have quite diversity
from different rounds.
# round1
Task QPS baseline StdDev QPS bsearc
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759069434
I don't remember any problem there. The toolchain use for
`generateJdkApiJarXX` was added much later and I did not touch that. Maybe
check git blame.
--
This is an automated m
dweiss commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759073292
I did check that file's history - it used an explicit exclusion for
compileMain19Java:
```
if (task.path == ":lucene:core:compileMain19Java") {
// uses "usc
gf2121 commented on issue #12659:
URL: https://github.com/apache/lucene/issues/12659#issuecomment-1759141669
I write a JMH benchmark to compare reading `vint` vs `msbvint`.
```
Benchmark (size) (valueBit) Mode Cnt Score
Error Units
ReadVLongBenchmar
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759182112
Hi, there are some other problem with the regenerate task, toolchains do not
automatically download, although enabled. I am digging.
--
This is an automated message from the Ap
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759187489
Very strange output if I clean the JDK cache in `~/.gradle/jdks`.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759191108
Task info by `gradlew tasks` is correct:
generateJdkApiJar19 - Regenerate the API-only JAR file with public Panama
Foreign & Vector API from JDK 19
generateJdkApiJar20 -
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759203200
> I did check that file's history - it used an explicit exclusion for
compileMain19Java:
>
> ```
> if (task.path == ":lucene:core:compileMain19Java") {
>
gf2121 commented on issue #12659:
URL: https://github.com/apache/lucene/issues/12659#issuecomment-1759244758
I tried another reading way like:
```
byte b = msbBytes[pos++];
long i = b & 0x7FL;
while (b < 0) {
b = msbBytes[pos++];
i = (i << 7) | (b & 0x7FL);
}
```
gf2121 opened a new pull request, #12661:
URL: https://github.com/apache/lucene/pull/12661
ISSUE: https://github.com/apache/lucene/issues/12659
--
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
gf2121 commented on issue #12659:
URL: https://github.com/apache/lucene/issues/12659#issuecomment-1759262343
Another possible reason I'm thinking is that maybe `readMSBVLong` is not as
hot as `readVLong` so compiler is not optimizing it enough, or the `readbyte`
in statistic context may con
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759281805
OK, I found the bug for the icorrect logger output, but that's not the issue
here. The fix is easy. I will provide a PR for the regenerator.
The reason is that it looks lik
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759287508
This fixes the logging problem:
```patch
diff --git a/gradle/generation/extract-jdk-apis.gradle
b/gradle/generation/extract-jdk-apis.gradle
index 28f56ea6432..020dc05
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759309453
It looks like gradle toolchain download is fully broken in Gradle 8.4. It
always says that it cannot find any JDK, no matter which OS (Windows, Linux).
--
This is an automated
uschindler opened a new pull request, #12662:
URL: https://github.com/apache/lucene/pull/12662
Fix the toolchain support when using 8.4:
- Logging of failures is wrong as in Groovy the for loop varaibale is not
final, wo when it is used in closure which is executed later you get the l
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759316894
Here is the draft PR, it only fixes the logging, but I have no idea why the
toolchains are not downloaded. If you clean `~/.gradle/jdks` and then run
`gradlew :lucene:core:regene
dweiss opened a new pull request, #12663:
URL: https://github.com/apache/lucene/pull/12663
As per discussion in: https://github.com/apache/lucene/issues/12641
This patch adds the ability to compile JMH microbenchmarks and run them from
within Lucene codebase. I didn't use any plugins
dweiss commented on issue #12641:
URL: https://github.com/apache/lucene/issues/12641#issuecomment-1759370475
I filed an initial take on this here:
https://github.com/apache/lucene/pull/12663
Copied just one of your existing benchmarks, Rob. Try if it fits your needs.
I mentioned on
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759372394
I found the issue, in Gradle 8+ you need to explicitely configure toolchain
repositories:
> [Using automatic toolchain downloading without having a repository
configured](
dweiss commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759389210
> because the loop variable is not final (like in Java)
In groovy closures do have access to their full context (including local
variables outside). That code used gstring, whi
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759397039
OK, PR is ready. The plugin needs to be in settings.gradle (it is a
so-called "settings" plugin, which works on gradle toplevel without project
context.
PR: #12662 (please
uschindler commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759401511
> > because the loop variable is not final (like in Java)
>
> In groovy closures do have access to their full context (including local
variables outside). That code used gs
uschindler merged PR #12662:
URL: https://github.com/apache/lucene/pull/12662
--
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.
uschindler closed issue #12655: Upgrade to Gradle 8.4
URL: https://github.com/apache/lucene/issues/12655
--
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:
benwtrent commented on code in PR #12657:
URL: https://github.com/apache/lucene/pull/12657#discussion_r1356732427
##
lucene/core/src/java/org/apache/lucene/codecs/lucene95/IncrementalHnswGraphMerger.java:
##
@@ -0,0 +1,226 @@
+/*
+ * Licensed to the Apache Software Foundation (A
benwtrent commented on code in PR #12657:
URL: https://github.com/apache/lucene/pull/12657#discussion_r1356733186
##
lucene/core/src/java/org/apache/lucene/codecs/lucene95/IncrementalHnswGraphMerger.java:
##
@@ -0,0 +1,226 @@
+/*
+ * Licensed to the Apache Software Foundation (A
benwtrent commented on code in PR #12657:
URL: https://github.com/apache/lucene/pull/12657#discussion_r1356734118
##
lucene/core/src/java/org/apache/lucene/codecs/lucene95/IncrementalHnswGraphMerger.java:
##
@@ -0,0 +1,226 @@
+/*
+ * Licensed to the Apache Software Foundation (A
jpountz commented on PR #12664:
URL: https://github.com/apache/lucene/pull/12664#issuecomment-1759536405
`luceneutil` on `wikibigall` gave good results, better than I expected:
```
TaskQPS baseline StdDevQPS
my_modified_version StdDev
jpountz commented on code in PR #12664:
URL: https://github.com/apache/lucene/pull/12664#discussion_r1356764107
##
lucene/core/src/java/org/apache/lucene/codecs/lucene90/Lucene90PostingsReader.java:
##
@@ -1143,14 +1151,16 @@ private void refillDocs() throws IOException {
risdenk commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759556593
Thanks @uschindler and @dweiss sorry for the work here :( I'll try to make
sure I test all these cases next time.
--
This is an automated message from the Apache Git Service.
To r
jpountz commented on code in PR #12664:
URL: https://github.com/apache/lucene/pull/12664#discussion_r1356769495
##
lucene/core/src/java/org/apache/lucene/codecs/MultiLevelSkipListReader.java:
##
@@ -113,13 +113,14 @@ public int skipTo(int target) throws IOException {
// w
rmuir commented on PR #12663:
URL: https://github.com/apache/lucene/pull/12663#issuecomment-1759558774
@dweiss thank you for this! I tried it out and it works great. Glad to see
the `cosineDistanceVectorUtil()` working in this example which is exactly what
I want to do.
--
This is an aut
jpountz commented on PR #12664:
URL: https://github.com/apache/lucene/pull/12664#issuecomment-1759583212
I tried to re-inline advance() but it still suggested a small slowdown for
`HighTermDayOfYearSort` and `CountAndHighMed` too so I suspect that there is
something else. There is indeed on
jpountz commented on PR #12664:
URL: https://github.com/apache/lucene/pull/12664#issuecomment-1759592840
Another potential reason for the slowdowns is that they're caused by the
changes to `MultiLevelSkipListReader`. Removing the additional checks would
require changing the way we encode po
mayya-sharipova commented on issue #11507:
URL: https://github.com/apache/lucene/issues/11507#issuecomment-1759594047
Yes, thanks for the reminder. Now Codec is responsible for managing dims, we
can close it.
--
This is an automated message from the Apache Git Service.
To respond to the m
mayya-sharipova closed issue #11507: Increase the number of dims for KNN
vectors to 2048 [LUCENE-10471]
URL: https://github.com/apache/lucene/issues/11507
--
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 g
dweiss commented on PR #12663:
URL: https://github.com/apache/lucene/pull/12663#issuecomment-1759605550
Let me review the minor remaining bits - I had to leave the computer before
the tests completed.
--
This is an automated message from the Apache Git Service.
To respond to the message,
benwtrent commented on PR #12663:
URL: https://github.com/apache/lucene/pull/12663#issuecomment-1759607611
@dweiss this is awesome!!! Thank you so much!
--
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
benwtrent commented on code in PR #12657:
URL: https://github.com/apache/lucene/pull/12657#discussion_r1356818110
##
lucene/core/src/java/org/apache/lucene/codecs/lucene95/IncrementalHnswGraphMerger.java:
##
@@ -0,0 +1,226 @@
+/*
+ * Licensed to the Apache Software Foundation (A
jpountz commented on code in PR #12658:
URL: https://github.com/apache/lucene/pull/12658#discussion_r1356844946
##
lucene/core/src/java/org/apache/lucene/codecs/PointsWriter.java:
##
@@ -202,7 +202,7 @@ public long size() {
@Override
public int ge
dweiss commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759671663
> By using a each closure the jdkVersion is private to the closure and the
build-time evaluation has its own jdkVersion variable (which is a constant).
Ah, thanks for clarifyin
dweiss commented on issue #12655:
URL: https://github.com/apache/lucene/issues/12655#issuecomment-1759674565
No worries, @risdenk - it's a big effort already, there'll always be
problems, especially with major version upgrades.
--
This is an automated message from the Apache Git Service.
jpountz commented on PR #12638:
URL: https://github.com/apache/lucene/pull/12638#issuecomment-1759681148
@yugushihuang Your explanation suggests that a `TermStates` could be created
somewhere and then used in a different context. But this is not how it's
expected to be used in general, the
shubhamvishu commented on PR #12548:
URL: https://github.com/apache/lucene/pull/12548#issuecomment-1759689286
Thanks for reviewing!
@benwtrent I have added a CHANGES.txt entry now, could you help me merging
this if all looks good?
--
This is an automated message from the Apache Gi
benwtrent commented on PR #12548:
URL: https://github.com/apache/lucene/pull/12548#issuecomment-1759701427
@shubhamvishu running CI :). I will see about merging and such soon.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub an
benwtrent commented on PR #12663:
URL: https://github.com/apache/lucene/pull/12663#issuecomment-1759752950
@dweiss one suggestion I have is adding a `README` for
`lucene/benchmark-jmh` on how to run the benchmarks and add new ones. Your
description on how to use it in this PR seems adequate
dweiss commented on PR #12663:
URL: https://github.com/apache/lucene/pull/12663#issuecomment-1759758762
I enabled forbiddenapis to the extent possible, to make Uwe happier. ;) I
like those flame-graphs that Solr has, etc., but these can come later. I'll add
a simple ```gradlew :helpJmh``` t
s1monw merged PR #12647:
URL: https://github.com/apache/lucene/pull/12647
--
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.apac
dweiss commented on PR #12663:
URL: https://github.com/apache/lucene/pull/12663#issuecomment-1759777296
@benwtrent Can you take a look? If there's any wording you think would work
better, please do change it. Explaining how JMH works is perhaps beyond the
scope of the readme file... ;)
--
benwtrent commented on PR #12663:
URL: https://github.com/apache/lucene/pull/12663#issuecomment-1759795868
Looks good @dweiss I added a line with an example for running a single
benchmark or suite of benchmarks.
--
This is an automated message from the Apache Git Service.
To respond to th
benwtrent commented on code in PR #12663:
URL: https://github.com/apache/lucene/pull/12663#discussion_r1356970803
##
lucene/benchmark-jmh/README.txt:
##
@@ -0,0 +1,21 @@
+The :lucene:benchmark-jmh module contains can be used to compile
+and execute JMH (https://github.com/openjd
shubhamvishu commented on PR #12636:
URL: https://github.com/apache/lucene/pull/12636#issuecomment-1759814356
@benwtrent Do you mean we earlier had similar implementation or such
interface?
> we decided to switch it as a common interface required either
Which interface are you
msokolov commented on code in PR #12548:
URL: https://github.com/apache/lucene/pull/12548#discussion_r1357027508
##
lucene/core/src/java/org/apache/lucene/search/DoubleValuesSource.java:
##
@@ -172,6 +173,52 @@ public LongValuesSource rewrite(IndexSearcher searcher)
throws IOEx
msokolov commented on PR #12548:
URL: https://github.com/apache/lucene/pull/12548#issuecomment-1759887696
oh I missed that @benwtrent had approved -- it's OK w/me to merge as-is,
just had some idea about organizing the static methods into a different class...
--
This is an automated messa
easyice commented on code in PR #12658:
URL: https://github.com/apache/lucene/pull/12658#discussion_r1357032536
##
lucene/core/src/java/org/apache/lucene/util/bkd/BKDWriter.java:
##
@@ -1248,7 +1259,8 @@ private void writeIndex(
metaOut.writeBytes(maxPackedValue, 0, config.
shubhamvishu commented on PR #12548:
URL: https://github.com/apache/lucene/pull/12548#issuecomment-1759892123
> oh I missed that @benwtrent had approved -- it's OK w/me to merge as-is,
just had some idea about organizing the static methods into a different class...
Thanks @msokolov !
easyice commented on code in PR #12658:
URL: https://github.com/apache/lucene/pull/12658#discussion_r1357049989
##
lucene/core/src/java/org/apache/lucene/codecs/PointsWriter.java:
##
@@ -202,7 +202,7 @@ public long size() {
@Override
public int ge
shubhamvishu commented on code in PR #12548:
URL: https://github.com/apache/lucene/pull/12548#discussion_r1357050474
##
lucene/core/src/java/org/apache/lucene/search/DoubleValuesSource.java:
##
@@ -172,6 +173,52 @@ public LongValuesSource rewrite(IndexSearcher searcher)
throws
msokolov commented on issue #12579:
URL: https://github.com/apache/lucene/issues/12579#issuecomment-1759925268
Interesting - so this is kind of like a noisy radius search in high
dimensions? It makes sense to me intuitively since we don't generally expect
searches to have the same number of
easyice commented on code in PR #12658:
URL: https://github.com/apache/lucene/pull/12658#discussion_r1357058015
##
lucene/core/src/java/org/apache/lucene/util/bkd/BKDWriter.java:
##
@@ -519,9 +526,8 @@ private Runnable writeFieldNDims(
// compute the min/max for this slice
easyice commented on code in PR #12658:
URL: https://github.com/apache/lucene/pull/12658#discussion_r1357058015
##
lucene/core/src/java/org/apache/lucene/util/bkd/BKDWriter.java:
##
@@ -519,9 +526,8 @@ private Runnable writeFieldNDims(
// compute the min/max for this slice
shubhamvishu commented on code in PR #12548:
URL: https://github.com/apache/lucene/pull/12548#discussion_r1357050474
##
lucene/core/src/java/org/apache/lucene/search/DoubleValuesSource.java:
##
@@ -172,6 +173,52 @@ public LongValuesSource rewrite(IndexSearcher searcher)
throws
gsmiller commented on code in PR #12642:
URL: https://github.com/apache/lucene/pull/12642#discussion_r1357072892
##
lucene/facet/src/test/org/apache/lucene/facet/TestDrillSideways.java:
##
@@ -316,6 +316,58 @@ public void testBasic() throws Exception {
IOUtils.close(searche
easyice commented on PR #12658:
URL: https://github.com/apache/lucene/pull/12658#issuecomment-1759980060
> I don't like when merges allocate O(maxDoc) memory so I'm keen to fixing
this. The idea looks correct to me, I wonder if we can make it a bit more
robust?
okay, I will take a lo
risdenk opened a new pull request, #2681:
URL: https://github.com/apache/lucene-solr/pull/2681
(no comment)
--
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,
gsmiller commented on code in PR #12642:
URL: https://github.com/apache/lucene/pull/12642#discussion_r1357145658
##
lucene/facet/src/test/org/apache/lucene/facet/TestDrillSideways.java:
##
@@ -316,6 +317,68 @@ public void testBasic() throws Exception {
IOUtils.close(searche
gf2121 commented on code in PR #12642:
URL: https://github.com/apache/lucene/pull/12642#discussion_r1357151214
##
lucene/facet/src/test/org/apache/lucene/facet/TestDrillSideways.java:
##
@@ -316,6 +316,58 @@ public void testBasic() throws Exception {
IOUtils.close(searcher.
benwtrent commented on PR #12548:
URL: https://github.com/apache/lucene/pull/12548#issuecomment-1760058838
These are good ideas @msokolov , @shubhamvishu a follow up PR is most
welcome.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on t
benwtrent closed issue #12394: Add the ability to compute vector similarity
scores with the new ValuesSource API
URL: https://github.com/apache/lucene/issues/12394
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL a
benwtrent merged PR #12548:
URL: https://github.com/apache/lucene/pull/12548
--
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
mayya-sharipova commented on code in PR #12582:
URL: https://github.com/apache/lucene/pull/12582#discussion_r1357182135
##
lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99HnswVectorsFormat.java:
##
@@ -0,0 +1,209 @@
+/*
+ * Licensed to the Apache Software Foundati
benwtrent commented on PR #12657:
URL: https://github.com/apache/lucene/pull/12657#issuecomment-1760088558
@zhaih I updated the API a bit. This is more like I was thinking. Having a
builder that accepts readers, doc maps, etc. And then can build with the final
merge state.
--
This is an
msokolov commented on code in PR #12548:
URL: https://github.com/apache/lucene/pull/12548#discussion_r1357190877
##
lucene/core/src/java/org/apache/lucene/search/DoubleValuesSource.java:
##
@@ -172,6 +173,52 @@ public LongValuesSource rewrite(IndexSearcher searcher)
throws IOEx
benwtrent commented on PR #12636:
URL: https://github.com/apache/lucene/pull/12636#issuecomment-1760104677
> Which interface are you referring to RandomAccessVectorValues? as that
uses generics to avoid duplication?
During the refactoring for byte/float I did a while back, during
msokolov commented on PR #12660:
URL: https://github.com/apache/lucene/pull/12660#issuecomment-1760118913
This looks like a great start. I worked up a very similar PR but I think I
have some concurrency bugs around the update/add of new entry points in the
upper graph levels. I might post m
s1monw commented on code in PR #12622:
URL: https://github.com/apache/lucene/pull/12622#discussion_r1357159891
##
lucene/core/src/java/org/apache/lucene/index/MergePolicy.java:
##
@@ -288,11 +288,32 @@ final void close(
}
}
-/** Wrap the reader in order to add/
s1monw commented on PR #12622:
URL: https://github.com/apache/lucene/pull/12622#issuecomment-1760126316
just for the record. I think we should record if a segment was written and
it contains blocks and make it viral. that might be a good idea anyway.
--
This is an automated message from
dweiss merged PR #12663:
URL: https://github.com/apache/lucene/pull/12663
--
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.apac
dweiss closed issue #12641: ability to run JMH benchmarks from gradle
URL: https://github.com/apache/lucene/issues/12641
--
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 unsu
dweiss commented on issue #12641:
URL: https://github.com/apache/lucene/issues/12641#issuecomment-1760149489
I left just a single benchmark in, @rmuir - if you'd like to move all of
them, please do so. If something doesn't work, let me know (it should though).
--
This is an automated mess
dweiss commented on code in PR #12663:
URL: https://github.com/apache/lucene/pull/12663#discussion_r1357235935
##
lucene/benchmark-jmh/README.txt:
##
@@ -0,0 +1,21 @@
+The :lucene:benchmark-jmh module contains can be used to compile
+and execute JMH (https://github.com/openjdk/j
zhaih commented on code in PR #12651:
URL: https://github.com/apache/lucene/pull/12651#discussion_r1357260551
##
lucene/core/src/java/org/apache/lucene/util/hnsw/HnswGraphBuilder.java:
##
@@ -288,7 +294,6 @@ private void selectAndLinkDiverse(
// only adding it if it is cl
zhaih commented on code in PR #12651:
URL: https://github.com/apache/lucene/pull/12651#discussion_r1357261063
##
lucene/core/src/java/org/apache/lucene/util/hnsw/HnswGraphSearcher.java:
##
@@ -232,7 +231,6 @@ void searchLevel(
graphSeek(graph, level, topCandidateNode);
zhaih commented on code in PR #12651:
URL: https://github.com/apache/lucene/pull/12651#discussion_r1357264961
##
lucene/core/src/java/org/apache/lucene/util/hnsw/OnHeapHnswGraph.java:
##
@@ -40,31 +41,29 @@ public final class OnHeapHnswGraph extends HnswGraph
implements Account
jpountz opened a new issue, #12665:
URL: https://github.com/apache/lucene/issues/12665
### Description
It would be nice to enable recursive graph bisection out of the box, so that
users don't even have to know that it exists or what it is to enjoy its
search-time performance benefits
zhaih commented on PR #12651:
URL: https://github.com/apache/lucene/pull/12651#issuecomment-1760235738
OK I have incorporate all the learning I have from #12660 and added several
more assertions to make it safer, please take a look again when you have time
@msokolov, thanks!
--
This is a
shubhamvishu commented on code in PR #12548:
URL: https://github.com/apache/lucene/pull/12548#discussion_r1357312134
##
lucene/core/src/java/org/apache/lucene/search/DoubleValuesSource.java:
##
@@ -172,6 +173,52 @@ public LongValuesSource rewrite(IndexSearcher searcher)
throws
rmuir commented on issue #12641:
URL: https://github.com/apache/lucene/issues/12641#issuecomment-1760262080
@dweiss thank you, I will fix and cleanup all the benchmarks and archive
https://github.com/rmuir/vectorbench as soon as I can :)
--
This is an automated message from the Apache Git
rmuir commented on issue #12641:
URL: https://github.com/apache/lucene/issues/12641#issuecomment-1760263361
I can also add some stuff from the README there such as how to install the
disassembler. It is stuff that you forget easily if you haven't done it in a
while.
--
This is an automat
zhaih commented on PR #12660:
URL: https://github.com/apache/lucene/pull/12660#issuecomment-1760280351
That sounds great, thanks Mike! Please post the PR and I will try to see
how to put it together. Meanwhile let's try to get the prerequisite PR
reviewed and pushed first? Such that thi
Bluetopia opened a new issue, #12666:
URL: https://github.com/apache/lucene/issues/12666
### Description
Using:
https://javadoc.io/doc/org.apache.lucene/lucene-core/latest/org/apache/lucene/analysis/Analyzer.html
which should hopefully be up to date. The current state of the source f
mayya-sharipova commented on code in PR #12582:
URL: https://github.com/apache/lucene/pull/12582#discussion_r1357345297
##
lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99ScalarQuantizedVectorsWriter.java:
##
@@ -0,0 +1,782 @@
+/*
+ * Licensed to the Apache Softwa
benwtrent commented on code in PR #12657:
URL: https://github.com/apache/lucene/pull/12657#discussion_r1357376455
##
lucene/core/src/java/org/apache/lucene/codecs/lucene95/IncrementalHnswGraphMerger.java:
##
@@ -0,0 +1,209 @@
+/*
+ * Licensed to the Apache Software Foundation (A
benwtrent commented on code in PR #12582:
URL: https://github.com/apache/lucene/pull/12582#discussion_r1357380984
##
lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99HnswVectorsFormat.java:
##
@@ -0,0 +1,209 @@
+/*
+ * Licensed to the Apache Software Foundation (AS
rmuir opened a new pull request, #12667:
URL: https://github.com/apache/lucene/pull/12667
Following up to @dweiss work, this gives us the same benchmarks as
https://github.com/rmuir/vectorbench, just without the code duplication and
maintenance hassle.
Each method is simply invoked w
rmuir commented on PR #12667:
URL: https://github.com/apache/lucene/pull/12667#issuecomment-1760526957
You can still tell what's happening too due to the log messages. When each
benchmark runs, you see a single message:
xxxScalar() methods:
```
WARNING: Java vector incubator mod
dungba88 commented on code in PR #12624:
URL: https://github.com/apache/lucene/pull/12624#discussion_r1353826324
##
lucene/core/src/java/org/apache/lucene/util/fst/BytesStore.java:
##
@@ -21,19 +21,18 @@
import java.util.List;
import org.apache.lucene.store.DataInput;
import
zhaih commented on code in PR #12651:
URL: https://github.com/apache/lucene/pull/12651#discussion_r1357264961
##
lucene/core/src/java/org/apache/lucene/util/hnsw/OnHeapHnswGraph.java:
##
@@ -40,31 +41,29 @@ public final class OnHeapHnswGraph extends HnswGraph
implements Account
1 - 100 of 110 matches
Mail list logo
