Skip to content

issues Search Results · repo:datastax/jvector language:Java

Filter by

138 results
 (45 ms)

138 results

indatastax/jvector (press backspace or delete to remove)

datastax/jvector
Should we consider supporting sparse vector?

pg-vector/nmslib and other projects support sparse vector capabilities. I took a quick look at their implementation principles: they did not modify the implementation of hnsw, but only supported sparse ...
  • xjtushilei
  • Opened 
    18 days ago
  • #401

datastax/jvector
Test TestOnDiskGraphIndex.testReorderingWithHoles fails

In line 182, onDiskGraph.getIdUpperBound() returns the size (3), when it should return the first ordinal greater than all node ids in the graph, i.e., 11.
bug
  • marianotepper
  • 1
  • Opened 
    26 days ago
  • #399

datastax/jvector
Test2DThreshold failures

The tests in Test2DThreshold have been behaving erratically
  • marianotepper
  • 1
  • Opened 
    on Jan 23
  • #390

datastax/jvector
AssertionError in testNVQEncodings

java.lang.AssertionError: 0.004292624942636493 0.002 for DOT_PRODUCT with 256 dimensions and 1 subvectors reproduces close to 100% of the time for me
  • jbellis
  • Opened 
    on Jan 9
  • #385

datastax/jvector
Reenable usage of SimdOps.assembleAndSum. Implement Panama/Native equivalent for CosineDecoder acceleration

We disabled SimdOps.assembleAndSum from PanamaVectorUtilSupport.assembleAndSum due to https://bugs.openjdk.org/browse/JDK-8321215. Now that released JDKs are fixed, we should re-test this in Bench and ...
  • jkni
  • 3
  • Opened 
    on Nov 12, 2024
  • #367

datastax/jvector
Interesting approach to RAG and vector db for long context

Please enable discussions for this GitHub project, the below is not really an issue: https://www.zyphra.com/post/reaching-1b-context-length-with-rag
  • hrstoyanov
  • 2
  • Opened 
    on Oct 21, 2024
  • #365

datastax/jvector
Usage of imprecise fp-model=fast.

With ANN search, we accept giving up accuracy for speed. Since most of the code in jvector_simd.c deals in floating-point computations, it may make sense to pass—fp-model=fast to the GCC compiler at build ...
  • jbhateja
  • 1
  • Opened 
    on Sep 19, 2024
  • #360

datastax/jvector
Cosine metric calculation uses different precision in default provider than in Panama and native

Cosine metric calculation uses different precision in final calculation in default and Panama/native providers so they produce slightly different results. Usually that should not be important, but we had ...
  • k-jamroz
  • 4
  • Opened 
    on Sep 10, 2024
  • #357

datastax/jvector
RandomAccessScoreProvider with MapRandomAccessVectorValues + non-sequential IDs produces wrong centroid

In 97e523c306ae42c3e963484e320fa1c7432b5250 approximateCentroid() implementation for the BuildScoreProvider returned from BuildScoreProvider.randomAccessScoreProvider() was updated to allow for non-sequential ...
  • vbekiaris
  • 3
  • Opened 
    on Aug 14, 2024
  • #354

datastax/jvector
jvector supports filter search

Does jvector supports filter search? I didn t find code related to filter search, such as metadata filtering
  • hanqiushi
  • 1
  • Opened 
    on Jul 29, 2024
  • #348
Issue origami icon

Learn how you can use GitHub Issues to plan and track your work.

Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub Issues
ProTip! 
Restrict your search to the title by using the in:title qualifier.
Issue origami icon

Learn how you can use GitHub Issues to plan and track your work.

Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub Issues
ProTip! 
Restrict your search to the title by using the in:title qualifier.
Issue search results · GitHub