[Save this search]

Status
All
   Fixed (7187)
  Closed (4390)
✓Open (2519)
   Won't Fix (543)
   Duplicate (297)
   Invalid (217)
   Not A Problem (195)
Issue type
✓All
  Issue (2225)
  PR (294)
Author relation
✓All
  None (2032)
  Member (595)
  Contributor (398)
  New contributor (53)
Created
✓All
  Past day (1)
  Past 2 days (7)
  Past 3 days (9)
  Past week (20)
  Past month (49)
  Past 3 months (78)
  Past 6 months (152)
  Past year (280)
Updated
✓All
  Past day (9)
  Past 2 days (27)
  Past 3 days (31)
  Past week (42)
  Past month (74)
  Past 3 months (112)
  Past 6 months (390)
  Past year (500)
Updated ago
✓All
  > 1 day ago (2510)
  > 2 days ago (2492)
  > 3 days ago (2488)
  > 1 week ago (2477)
  > 1 month ago (2445)
  > 3 months ago (2407)
  > 1 year ago (2019)
Comment count
✓All
  0 (556)
  1 (352)
  2 - 5 (789)
  6 - 10 (427)
  10 - 20 (302)
  > 20 (162)
Reaction count
✓All
  0 (2240)
  1 (166)
  2 - 5 (91)
  6 - 10 (16)
  10 - 20 (5)
  > 20 (1)
Review Requested
✓All
  jpountz (14)
  mikemccand (8)
  benwtrent (5)
  dweiss (4)
  iverase (4)
  rmuir (3)
  msokolov (3)

See all 22...
Mentioned
✓All
  jpountz (89)
  mikemccand (70)
  rmuir (50)
  uschindler (38)
  benwtrent (38)
  msokolov (37)
  dweiss (17)

See all 195...
Reviewed
✓All
  jpountz (23)
  mikemccand (23)
  dweiss (12)
  uschindler (12)
  benwtrent (12)
  rmuir (11)
  msokolov (10)

See all 58...
Commented
✓All
  asfimport (1369)
  github-actions[bot] (222)
  jpountz (140)
  mikemccand (115)
  rmuir (98)
  benwtrent (76)
  msokolov (75)

See all 282...
User
✓All
  asfimport (1811)
  github-actions[bot] (261)
  mikemccand (199)
  jpountz (198)
  rmuir (128)
  benwtrent (116)
  msokolov (94)

See all 430...
Last comment user
✓All
  asfimport (1326)
  github-actions[bot] (216)
  jpountz (32)
  mikemccand (28)
  rmuir (24)
  benwtrent (21)
  dweiss (16)

See all 152...
Draft
✓All
  No (241)
  Yes (53)
Component
✓All
  core (583)
  analysis (149)
  highlighter (46)
  spatial (41)
  facet (39)
  queryparser (26)
  test-framework (24)

See all 23...
Type
✓All
  enhancement (1158)
  bug (726)
  task (197)
  test (77)
  documentation (22)
Labels
✓All
  legacy-jira-fix-versio... (214)
  Stale (212)
  legacy-jira-fix-versio... (169)
  affects-version:4.0-ALPHA (81)
  tool:build (54)
  vector-based-search (47)
  affects-version:6.0 (37)

See all 157...
Commits?
✓All
  No (2519)
Reporter
✓All
  rmuir (271)
  mikemccand (153)
  jpountz (126)
  dsmiley (68)
  uschindler (52)
  iverase (41)
  dweiss (40)

See all 733...
Assignee
✓All
  Unassigned (2278)
  mikemccand (35)
  uschindler (31)
  romseygeek (27)
  dsmiley (22)
  rmuir (14)
  jpountz (11)

See all 44...
  
Filters: Status (Open),  Issue type,  Author relation,  Created,  Updated,  Updated ago,  Comment count,  Reaction count,  Review Requested,  Mentioned,  Reviewed,  Commented,  User,  Last comment user,  Draft,  Component,  Type,  Labels,  Commits?,  Reporter,  Assignee

#15676 PR: Feature/collaborative hnsw search
30.9 minutes ago  4 comments  0 votes  0 watches  benwtrent, github-actions[bot], krickert, navneet1v
[HNSW] Collaborative Search via Dynamic Threshold Feedback > **Status: Experimental** - This ... Full distributed multi-shard testing will happen in a follow-up PR for OpenSearch.
    krickert 30.9 minutes ago:   Refactored Implementation: Floor-Score Sharing & Real-World Scaling Analysis Thanks for the ... I've refactored the CollaborativeKnnCollector to share the k-th best score (floor) instead of the ... Key Change: Floor-Score Accumulation The previous implementation accumulated every collected doc's ...
    navneet1v 2.6 hours ago:  Hi @krickert This is an interesting idea, and I think it extends current implementation of ... . +1 on the idea.

#15623 PR: deps(java): bump gradle-wrapper from 9.2.1 to 9.3.0
44.8 minutes ago  7 comments  0 votes  0 watches  dependabot, dependabot[bot], dweiss, uschindler
Bumps gradle-wrapper from 9.2.1 to 9.3.0. ![Dependabot compatibility score](https://docs.github.com/en/github/managing-security-vulnerabilities ... You can also trigger a rebase manually by commenting @dependabot rebase. [//]: (dependabot-automerg ...
    dweiss 44.8 minutes ago:  Yes. Not working for me though - benchmarks-jmh still shows no tests even though they're clearly there. I'll investigate when I have some time but I don't think it's our fault.
    uschindler 11.8 hours ago:  Did they say it's fixed?

#15612: Cluster Based ANN Vector Search for Lucene
1.7 hours ago  20 comments  0 votes  0 watches  Vikasht34, atris, benwtrent, navneet1v, vigyasharma
Our current HNSW vector search is great with low latency and high recall, but comes with its own ... Graph based algos are memory intensive which make them hard to scale. And they don't work well with segment merge.
    atris 1.7 hours ago:  Currently, the PR post-filters search results during the fine phase, scanning allowed vectors ... This avoids zero results for populated partitions but remains sensitive to disjoint query/filter ... Pruning via minSimilarity (and integrating with/emulating #15676) is a planned optimization to ...
    navneet1v 2.4 hours ago:  Hi @vigyasharma , @atris Thanks for adding this idea. I see that on this thread we have been discussing mainly on indexing, I would like to add some ... Similar to current HNSW search implementation we should add the concept of minSimilarity score to ...

#15662: Lucene nightly benchmark regression on 29/01
5.6 hours ago  38 comments  0 votes  0 watches  benwtrent, epotyom, mikemccand, msokolov, rmuir, scampi, shubhamsrkdev, stefanvodita, uschindler
Description We are seeing regressions on lucene nightly benchmarks on 29/01 The set of commits ... We should find the root cause of the regression. Version and environment details_No response_
    benwtrent 5.6 hours ago:  > And I think this means we shouldn't hold Lucene 10.4.0 release? Agreed! We haven't seen regressions on any of the other benchmarks, nor in the (albeit few) benchmarks we ...
    uschindler 5.7 hours ago:  > Yeah, +1, I'll make this switch after surviving the current situation once dust settles ... it'd ... Hmm we need a day when lucene devs are gonna be idle ... I think the biggest problem on the last change was the long "downtime" of the benchmark.

#15667 PR: Use bulk scoring in NeighborArray#isWorstNonDiverse
8.3 hours ago  17 comments  0 votes  0 watches  ML-dev-crypto, benwtrent, github-actions[bot], navneet1v
This change updates NeighborArrayisWorstNonDiverse to use the RandomVectorScorerbulkScore API, ... The bulkScore method is a default interface method and safely falls back to per-node scoring when ... Fixes 15606
    ML-dev-crypto 8.3 hours ago:  > @ML-dev-crypto do we know what is the impact in performance because of this change? Thanks for the question. Theoretically, this change eliminates array allocations in the hot path of graph construction and ...
    navneet1v 13.4 hours ago:  @ML-dev-crypto do we know what is the impact in performance because of this change?

#15027 PR: Optimize prefix sum computation in Lucene99HnswVectorsReader; fixes #15024
20.2 hours ago  18 comments  0 votes  0 watches  benwtrent, github-actions[bot], jpountz, msokolov, yossev
Replaced the two-step prefix sum loop in Lucene99HnswVectorsReader with a single-loop variant that ... Previous approach: - Read first value separately.
    github-actions[bot] 2.7 months ago:  This PR has not had activity in the past 2 weeks, labeling it as stale. If the PR is waiting for review, notify the dev@lucene.apache.org list. Thank you for your contribution!
    benwtrent 3.3 months ago:  Hello @yossev friendly ping :). Once conflicts & formatting are fixed, I can merge and backport. We should also have a CHANGES entry under optimizations for 10.4.0

#15177 PR: Update code comment after bumping block size to 256.
20.2 hours ago  2 comments  0 votes  0 watches  RamakrishnaChilaka, benwtrent, github-actions[bot], jpountz
The flag we store at the beginning of every block is either 0 if the block is dense, a positive ... This PR updates the comment that explains why we're guaranteed to not overflow a byte with the ...
    github-actions[bot] 3.8 months ago:  This PR has not had activity in the past 2 weeks, labeling it as stale. If the PR is waiting for review, notify the dev@lucene.apache.org list. Thank you for your contribution!

[20.6 msec search, 21.6 msec total]