original-brownbear commented on code in PR #12453:
URL: https://github.com/apache/lucene/pull/12453#discussion_r1271914546
##
lucene/core/src/java/org/apache/lucene/store/BufferedIndexInput.java:
##
@@ -159,6 +159,63 @@ public final long readLong() throws IOException {
}
jpountz opened a new issue, #12456:
URL: https://github.com/apache/lucene/issues/12456
While disjunctive queries got a performance boost with
https://github.com/apache/lucene/pull/12444
([OrHighHigh](http://people.apache.org/~mikemccand/lucenebench/OrHighHigh.html),
[OrHighMed](http://peop
jpountz commented on code in PR #12446:
URL: https://github.com/apache/lucene/pull/12446#discussion_r1271961619
##
lucene/core/src/java/org/apache/lucene/search/MaxScoreBulkScorer.java:
##
@@ -168,7 +171,17 @@ private boolean partitionScorers() {
if (maxScoreSumFloat >= m
jpountz commented on code in PR #12446:
URL: https://github.com/apache/lucene/pull/12446#discussion_r1271963742
##
lucene/core/src/java/org/apache/lucene/search/BulkScorer.java:
##
@@ -90,4 +90,13 @@ public abstract int score(LeafCollector collector, Bits
acceptDocs, int min, i
jpountz commented on code in PR #12453:
URL: https://github.com/apache/lucene/pull/12453#discussion_r1271956372
##
lucene/core/src/test/org/apache/lucene/store/TestBufferedIndexInput.java:
##
@@ -209,6 +209,103 @@ public void testBackwardsLongReads() throws IOException {
original-brownbear commented on code in PR #12453:
URL: https://github.com/apache/lucene/pull/12453#discussion_r1272006226
##
lucene/core/src/test/org/apache/lucene/store/TestBufferedIndexInput.java:
##
@@ -209,6 +209,103 @@ public void testBackwardsLongReads() throws IOExceptio
jpountz opened a new pull request, #12457:
URL: https://github.com/apache/lucene/pull/12457
Partitioning scorers is an optimization problem: the optimal set of
non-essential scorers is the subset of scorers whose sum of max window scores
is less than the minimum competitive score that maxim
jpountz commented on PR #12457:
URL: https://github.com/apache/lucene/pull/12457#issuecomment-1647837090
luceneutil doesn't show an improvement on wikimedium because all fuzzy
queries only have low-frequency terms, only nightlies have the `titel~2` query
in their tasks file.
```
jpountz commented on code in PR #12453:
URL: https://github.com/apache/lucene/pull/12453#discussion_r1272210731
##
lucene/CHANGES.txt:
##
@@ -84,6 +84,8 @@ Optimizations
* GITHUB#12372: Reduce allocation during HNSW construction (Jonathan Ellis)
+* GITHUB#12453: Faster bulk
jpountz commented on code in PR #12453:
URL: https://github.com/apache/lucene/pull/12453#discussion_r1272352389
##
lucene/CHANGES.txt:
##
@@ -84,6 +84,8 @@ Optimizations
* GITHUB#12372: Reduce allocation during HNSW construction (Jonathan Ellis)
+* GITHUB#12453: Faster bulk
jpountz merged PR #12442:
URL: https://github.com/apache/lucene/pull/12442
--
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.apa
HoustonPutman commented on PR #12430:
URL: https://github.com/apache/lucene/pull/12430#issuecomment-1648042240
This same fix worked for Solr:
https://solr.apache.org/docs/9_3_0/core/index.html
Will go ahead and merge!
--
This is an automated message from the Apache Git Service.
To
HoustonPutman merged PR #12430:
URL: https://github.com/apache/lucene/pull/12430
--
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...@luce
original-brownbear commented on code in PR #12453:
URL: https://github.com/apache/lucene/pull/12453#discussion_r1272393881
##
lucene/CHANGES.txt:
##
@@ -84,6 +84,8 @@ Optimizations
* GITHUB#12372: Reduce allocation during HNSW construction (Jonathan Ellis)
+* GITHUB#12453:
jpountz merged PR #12453:
URL: https://github.com/apache/lucene/pull/12453
--
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.apa
gsmiller commented on issue #12451:
URL: https://github.com/apache/lucene/issues/12451#issuecomment-1648287605
Also, here's the compiled automaton resulting from the code-point automaton
above:
jbellis commented on code in PR #12421:
URL: https://github.com/apache/lucene/pull/12421#discussion_r1272560387
##
lucene/core/src/java/org/apache/lucene/util/hnsw/HnswGraphSearcher.java:
##
@@ -311,7 +369,6 @@ void searchLevel(
graphSeek(graph, level, topCandidateNode);
jbellis commented on code in PR #12421:
URL: https://github.com/apache/lucene/pull/12421#discussion_r1272561099
##
lucene/core/src/java/org/apache/lucene/util/hnsw/ConcurrentHnswGraphBuilder.java:
##
@@ -0,0 +1,465 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) unde
benwtrent commented on code in PR #12421:
URL: https://github.com/apache/lucene/pull/12421#discussion_r1272563613
##
lucene/core/src/java/org/apache/lucene/util/hnsw/HnswGraphSearcher.java:
##
@@ -311,7 +369,6 @@ void searchLevel(
graphSeek(graph, level, topCandidateNode)
gsmiller opened a new issue, #12458:
URL: https://github.com/apache/lucene/issues/12458
### Description
When converting a unicode (UTF32) automaton down to a UTF8 representation,
UTF32toUTF8 can create an automaton that produces/accepts invalid UTF8. This
happens when a transition in
gsmiller commented on issue #12451:
URL: https://github.com/apache/lucene/issues/12451#issuecomment-1648504863
This looks like a bug in the `UTF32toUTF8` conversion logic to me. It seems
the resulting UTF8 automaton in this case is producing a large range of invalid
UTF8. This appears to al
benwtrent commented on code in PR #12434:
URL: https://github.com/apache/lucene/pull/12434#discussion_r1272680226
##
lucene/core/src/java/org/apache/lucene/util/hnsw/KnnResults.java:
##
@@ -0,0 +1,136 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
benwtrent commented on code in PR #12434:
URL: https://github.com/apache/lucene/pull/12434#discussion_r1272681200
##
lucene/core/src/java/org/apache/lucene/util/hnsw/KnnResultsProvider.java:
##
@@ -0,0 +1,25 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one o
benwtrent commented on code in PR #12434:
URL: https://github.com/apache/lucene/pull/12434#discussion_r1272735873
##
lucene/core/src/java/org/apache/lucene/util/hnsw/KnnResultsProvider.java:
##
@@ -0,0 +1,25 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one o
iverase opened a new issue, #12459:
URL: https://github.com/apache/lucene/issues/12459
### Description
Binary doc values allow to store a variable number of bytes on a doc value.
In order to read those bytes, we currently get a BytesRef from the API which
contains the bytes on heap.
iverase opened a new pull request, #12460:
URL: https://github.com/apache/lucene/pull/12460
see https://github.com/apache/lucene/issues/12459
--
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
iverase commented on issue #12459:
URL: https://github.com/apache/lucene/issues/12459#issuecomment-1648979938
I wrote a prototype for this change here:
https://github.com/apache/lucene/pull/12460
--
This is an automated message from the Apache Git Service.
To respond to the message, pleas
jmazanec15 commented on issue #12342:
URL: https://github.com/apache/lucene/issues/12342#issuecomment-1649102549
> 🤦 yep!
> Here is with the higher max conn. Sort of better.
Right, I was thinking this might explain the recall descrepency for the
dotproduct score change (0.989 vs 0
searchivarius commented on issue #12342:
URL: https://github.com/apache/lucene/issues/12342#issuecomment-1649112930
Hi @jmazanec15 and @benwtrent : thanks a lot for testing. For higher recalls
(somewhat higher or lower than 0.8) transformation seem to lead to substantial
increase in latency
29 matches
Mail list logo
