rapidsai/raft issues and pull requests

#2494 - Disable hnswlib if not compiling shared library components of RAFT

Pull Request - State: open - Opened by divyegala 3 days ago
Labels: improvement, breaking, cpp, CMake

#2493 - [WIP] add telemetry

Pull Request - State: open - Opened by msarahan 3 days ago
Labels: improvement, non-breaking, 2 - In Progress, Build Time Improvement, Benchmarks

#2492 - [FEA] Support the cached and reusable `sparsity` mechanism

Issue - State: open - Opened by rhdong 3 days ago
Labels: feature request

#2491 - [BUG] pylibraft compilation errors with Cython 3.1 nightlies

Issue - State: open - Opened by jameslamb 3 days ago
Labels: bug

#2490 - enforce wheel size limits and README formatting in CI, put a ceiling on Cython dependency

Pull Request - State: open - Opened by jameslamb 4 days ago - 3 comments
Labels: improvement, non-breaking, python, ci

#2489 - [FEA] Remaining APIs to migrate from RAFT

Issue - State: closed - Opened by cjnolet 8 days ago
Labels: feature request

#2488 - Removing deprecated VSS APIs from RAFT

Pull Request - State: open - Opened by cjnolet 8 days ago
Labels: improvement, non-breaking, cpp, CMake, python

#2487 - NCCL clique initialization function

Pull Request - State: open - Opened by viclafargue 8 days ago - 1 comment
Labels: improvement, non-breaking, cpp

#2486 - Put a ceiling on cuda-python

Pull Request - State: closed - Opened by bdice 11 days ago - 1 comment
Labels: bug, non-breaking, python

#2485 - [FEA] Find a way to generate and validate test data without hardcoding for Lanczos solver

Issue - State: open - Opened by lowener 11 days ago
Labels: feature request

#2484 - [BUG] `raft::linalg::gemv` calls can lead to `CUBLAS_STATUS_INVALID_VALUE`

Issue - State: open - Opened by lowener 11 days ago
Labels: bug

#2484 - [BUG] `raft::linalg::gemv` calls can lead to `CUBLAS_STATUS_INVALID_VALUE`

Issue - State: open - Opened by lowener 11 days ago
Labels: bug

#2483 - [FEA] Add largest eigenvalues computation to Lanczos

Issue - State: open - Opened by lowener 11 days ago
Labels: feature request

#2483 - [FEA] Add largest eigenvalues computation to Lanczos

Issue - State: open - Opened by lowener 11 days ago
Labels: feature request

#2482 - Add breaking change workflow trigger

Pull Request - State: open - Opened by AyodeAwe 11 days ago
Labels: improvement, non-breaking

#2481 - [FEA] Lanczos solver v2

Pull Request - State: open - Opened by lowener 15 days ago - 6 comments
Labels: 3 - Ready for Review, feature request, non-breaking, cpp, CMake, python

#2480 - Don't presume pointers location infers usability.

Pull Request - State: closed - Opened by robertmaynard 15 days ago - 1 comment
Labels: bug, non-breaking, cpp

#2479 - Update to CCCL 2.7.0-rc2.

Pull Request - State: open - Opened by bdice 16 days ago
Labels: improvement, non-breaking, cpp, CMake

#2479 - Update to CCCL 2.7.0-rc2.

Pull Request - State: open - Opened by bdice 16 days ago
Labels: improvement, non-breaking, cpp, CMake

#2478 - Do not initialize the pinned mdarray at construction time

Pull Request - State: closed - Opened by achirkin 16 days ago - 5 comments
Labels: 3 - Ready for Review, improvement, breaking, cpp

#2477 - [FEA] Raft primitives reduce_by_key

Issue - State: open - Opened by jperez999 18 days ago
Labels: feature request

#2472 - devcontainer: replace `VAULT_HOST` with `AWS_ROLE_ARN`

Pull Request - State: closed - Opened by jjacobelli 27 days ago - 1 comment
Labels: improvement, non-breaking

#2470 - print sccache stats in builds

Pull Request - State: closed - Opened by jameslamb about 1 month ago - 1 comment
Labels: improvement, non-breaking, ci

#2469 - Allow compilation with CUDA 12.6.1

Pull Request - State: closed - Opened by robertmaynard about 1 month ago - 1 comment
Labels: bug, non-breaking, cpp

#2425 - Exclude any kernel symbol that uses cutlass

Pull Request - State: closed - Opened by robertmaynard 3 months ago - 1 comment
Labels: bug, non-breaking

#2425 - Exclude any kernel symbol that uses cutlass

Pull Request - State: closed - Opened by robertmaynard 3 months ago - 1 comment
Labels: bug, non-breaking

#2416 - Lanczos Solver

Pull Request - State: closed - Opened by aamijar 3 months ago - 2 comments
Labels: improvement, non-breaking, cpp, CMake, python

#2353 - [REVIEW] Add tfidf bm25

Pull Request - State: open - Opened by jperez999 5 months ago - 3 comments
Labels: enhancement, 3 - Ready for Review, feature request, non-breaking, cpp, CMake, Vector Search

#2317 - Correct initializer list order found by cuvs

Pull Request - State: closed - Opened by robertmaynard 6 months ago - 1 comment
Labels: bug, 3 - Ready for Review, non-breaking, cpp

#2317 - Correct initializer list order found by cuvs

Pull Request - State: closed - Opened by robertmaynard 6 months ago - 1 comment
Labels: bug, 3 - Ready for Review, non-breaking, cpp

#2277 - Benchmark CAGRA + DiskANN

Pull Request - State: closed - Opened by tarang-jain 7 months ago - 3 comments
Labels: improvement, non-breaking, cpp, CMake, python, ci

#2254 - Correct member initialization order

Pull Request - State: closed - Opened by robertmaynard 8 months ago - 1 comment
Labels: bug, 3 - Ready for Review, non-breaking, cpp

#2254 - Correct member initialization order

Pull Request - State: closed - Opened by robertmaynard 8 months ago - 1 comment
Labels: bug, 3 - Ready for Review, non-breaking, cpp

#2250 - When using CUDA 12.4+ use zstd compression

Pull Request - State: closed - Opened by robertmaynard 8 months ago
Labels: 5 - DO NOT MERGE, improvement, non-breaking, cpp, CMake

#2250 - When using CUDA 12.4+ use zstd compression

Pull Request - State: closed - Opened by robertmaynard 8 months ago
Labels: 5 - DO NOT MERGE, improvement, non-breaking, cpp, CMake

#2045 - Correct function that wasn't returning a value

Pull Request - State: closed - Opened by robertmaynard 12 months ago - 1 comment
Labels: bug, 3 - Ready for Review, non-breaking, cpp

#2045 - Correct function that wasn't returning a value

Pull Request - State: closed - Opened by robertmaynard 12 months ago - 1 comment
Labels: bug, 3 - Ready for Review, non-breaking, cpp

#1623 - Use rapids-cmake new parallel testing feature

Pull Request - State: closed - Opened by robertmaynard over 1 year ago - 1 comment
Labels: feature request, non-breaking, cpp, CMake, ci

#1623 - Use rapids-cmake new parallel testing feature

Pull Request - State: closed - Opened by robertmaynard over 1 year ago - 1 comment
Labels: feature request, non-breaking, cpp, CMake, ci

#1054 - Specify correct global targets names to raft_export

Pull Request - State: closed - Opened by robertmaynard almost 2 years ago - 1 comment
Labels: bug, 3 - Ready for Review, non-breaking, cpp, CMake

#1054 - Specify correct global targets names to raft_export

Pull Request - State: closed - Opened by robertmaynard almost 2 years ago - 1 comment
Labels: bug, 3 - Ready for Review, non-breaking, cpp, CMake

#686 - Restore changes from #653 and #655 and correct cmake component dependencies

Pull Request - State: closed - Opened by robertmaynard over 2 years ago - 5 comments
Labels: 3 - Ready for Review, feature request, non-breaking, cpp, CMake

#686 - Restore changes from #653 and #655 and correct cmake component dependencies

Pull Request - State: closed - Opened by robertmaynard over 2 years ago - 5 comments
Labels: 3 - Ready for Review, feature request, non-breaking, cpp, CMake

#100 - [REVIEW] Add Divye as codeowner

Pull Request - State: closed - Opened by JohnZed almost 4 years ago - 1 comment

#99 - [REVIEW] Adding canberra and chebyshev to distance types

Pull Request - State: closed - Opened by cjnolet almost 4 years ago - 3 comments
Labels: improvement, non-breaking, 2 - In Progress

#98 - [REVIEW] Adding inner product to DistanceType for max inner product searches

Pull Request - State: closed - Opened by cjnolet almost 4 years ago - 1 comment
Labels: 3 - Ready for Review

#97 - [FEA] Use RMM directly instead of adapters

Issue - State: closed - Opened by divyegala about 4 years ago - 8 comments
Labels: inactive-30d, inactive-90d

#96 - [FEA] lap_kernels.cu numeric_limits/infinity

Issue - State: open - Opened by ChuckHastings about 4 years ago - 4 comments
Labels: inactive-30d, inactive-90d

#95 - [REVIEW] Cholesky rank one update prim

Pull Request - State: closed - Opened by tfeher about 4 years ago - 1 comment

#94 - [REVIEW] Allow generic reductions for the map then reduce op

Pull Request - State: closed - Opened by tfeher about 4 years ago - 3 comments

#93 - [REVIEW] Incorporate Date/Nagi implementation of Hungarian Algorithm

Pull Request - State: closed - Opened by ChuckHastings about 4 years ago - 1 comment

#92 - Gpuciscripts clean and update

Pull Request - State: closed - Opened by msadang about 4 years ago - 2 comments
Labels: 3 - Ready for Review, improvement, non-breaking

#91 - [FEA] Add cuFFT helpers

Issue - State: open - Opened by zbjornson about 4 years ago - 6 comments
Labels: inactive-30d, inactive-90d

#90 - [FEA] Add getDeviceCapability function

Issue - State: open - Opened by zbjornson about 4 years ago - 2 comments
Labels: inactive-30d, inactive-90d

#89 - [DO NOT MERGE] updated clang to v11.0.0

Pull Request - State: closed - Opened by teju85 about 4 years ago - 5 comments
Labels: inactive-30d, inactive-90d

#88 - [FEA] Upgrade clang version to include support for cuda-11

Issue - State: closed - Opened by teju85 about 4 years ago - 3 comments
Labels: inactive-30d, inactive-90d

#87 - [BUG] Fix clang-tidy errors properly after removing `// NOLINT` tags from the offending lines

Issue - State: open - Opened by teju85 about 4 years ago - 2 comments
Labels: bug, inactive-30d, inactive-90d

#86 - [REVIEW] missing headers for newly moved prims

Pull Request - State: closed - Opened by divyegala about 4 years ago - 1 comment

#85 - [WIP] Fix clang tidy errors in raft

Pull Request - State: closed - Opened by teju85 about 4 years ago - 7 comments
Labels: inactive-30d, inactive-90d

#84 - [BUG] Fix clang-tidy errors

Issue - State: closed - Opened by teju85 about 4 years ago - 3 comments
Labels: bug, inactive-30d, inactive-90d

#83 - [FEA] Deploy raft globally

Issue - State: closed - Opened by afender about 4 years ago - 4 comments
Labels: inactive-30d

#82 - [gpuCI] Auto-merge branch-0.16 to branch-0.17 [skip ci]

Pull Request - State: closed - Opened by GPUtester about 4 years ago - 1 comment

#81 - [REVIEW] BUG Temporarily expose internal NCCL communicator

Pull Request - State: closed - Opened by seunghwak about 4 years ago - 1 comment

#80 - [gpuCI] Auto-merge branch-0.16 to branch-0.17 [skip ci]

Pull Request - State: closed - Opened by GPUtester about 4 years ago - 1 comment

#79 - [REVIEW] BUG remove debug print statements

Pull Request - State: closed - Opened by ChuckHastings about 4 years ago - 1 comment

#78 - [FEA] Update RAFT comms to use ncclSend/ncclRecv for P2P

Issue - State: closed - Opened by seunghwak about 4 years ago - 2 comments
Labels: inactive-30d, inactive-90d

#77 - [REVIEW] Adding CUB in dependencies

Pull Request - State: closed - Opened by divyegala about 4 years ago - 4 comments

#76 - [gpuCI] Auto-merge branch-0.16 to branch-0.17 [skip ci]

Pull Request - State: closed - Opened by GPUtester about 4 years ago - 1 comment

#75 - [BUG] mpi_comms comm_split does not split NCCL communicator

Issue - State: open - Opened by seunghwak about 4 years ago - 2 comments
Labels: bug, inactive-30d, inactive-90d

#74 - [REVIEW] BUG Bug fix in RAFT comm_split in std_comms

Pull Request - State: closed - Opened by seunghwak about 4 years ago - 1 comment

#73 - [REVIEW] Move DistanceType enum to RAFT from cuML

Pull Request - State: closed - Opened by wphicks about 4 years ago - 2 comments
Labels: 3 - Ready for Review

#72 - [gpuCI] Auto-merge branch-0.16 to branch-0.17 [skip ci]

Pull Request - State: closed - Opened by GPUtester about 4 years ago - 1 comment

#71 - [BUG] handle_t destructor throws exceptions

Issue - State: open - Opened by zbjornson about 4 years ago - 2 comments
Labels: bug, inactive-30d, inactive-90d

#70 - [REVIEW] Adding CUB to RAFT cmake

Pull Request - State: closed - Opened by divyegala about 4 years ago - 1 comment

#69 - [REVIEW] Updates for RMM being header only

Pull Request - State: closed - Opened by dantegd about 4 years ago

#68 - [REVIEW] Change the type of recvcounts & displs in allgatherv from size_t[] & int[] to size_t* & size_t*, respectively.

Pull Request - State: closed - Opened by seunghwak about 4 years ago - 1 comment

GitHub / rapidsai/raft issues and pull requests