rhdong comments

Results 72 comments of


                                            rhdong

[BUG] Brute force python save_load test fail randomly

I found one query distances with different to samples are very close occasionally: ```shell E array([[7356, 6315, 3579, 2860, 3790, 1747, 9880, 2584, 4067, 2456], E [7520, 6982, 5815, 929,...

[BUG] Brute force python save_load test fail randomly

The current conclusion is a random failure caused because the distances are too close. There's still a reminded suspect point as to why the precision looks like a downgrade. As...

Java: Fix for a crash situation when using different threads (#680)

/ok to test

Java: Fix for a crash situation when using different threads (#680)

Hi @cjnolet , would you like this PR could be squashed with https://github.com/rapidsai/cuvs/pull/695 ?

[Opt] Enforce the UT Coverity and add benchmark for `transpose`

- Reopen for https://github.com/rapidsai/raft/pull/2421 for wrongly forced push

[Opt] Enforce the UT Coverity and add benchmark for `transpose`

/ok to test

[Opt] Enforce the UT Coverity and add benchmark for `transpose`

/merge

[FEA] Filtered CAGRA to automatically use pre-filtered brute-force when the filter ratio reaches a particular theshold.

Does the issue mentioned here depend on the specific dataset possibly?

[FEA] Filtered CAGRA to automatically use pre-filtered brute-force when the filter ratio reaches a particular theshold.

> No specific dataset, but the core use-case is hybrid search, where Lucene or Milvus might be accepting a filter that has been constructed based on having done a prior...

[FEA] Filtered CAGRA to automatically use pre-filtered brute-force when the filter ratio reaches a particular theshold.

> Now that we are quite far into implementation, I think more and more that this idea could backfire on us. > > 1. At this moment we haven't identified...