RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.
matrix::select_k
in brute_force::knn call (#1463) @benfredraft::resources
across raft::random
(#1420) @cjnoletbranch-23.06
and branch-23.04
(#1403) @galipremsagardevice_vector_view
(#1384) @lowenermatrix::select_k
in brute_force::knn call (#1463) @benfredraft::resources
across raft::random
(#1420) @cjnoletbranch-23.06
and branch-23.04
(#1403) @galipremsagardevice_vector_view
(#1384) @lowenermatrix::select_k
in brute_force::knn call (#1463) @benfredraft::resources
across raft::random
(#1420) @cjnoletbranch-23.06
and branch-23.04
(#1403) @galipremsagardevice_vector_view
(#1384) @lowenermatrix::select_k
in brute_force::knn call (#1463) @benfredraft::resources
across raft::random
(#1420) @cjnoletbranch-23.06
and branch-23.04
(#1403) @galipremsagardevice_vector_view
(#1384) @lowenerraft-dask/meta.yaml
(#1637) @jakirkhamexit(-1)
(#1594) @benfreddask
and distributed
for development and fix merge_labels
test (#1574) @galipremsagardask
and distributed
for release (#1399) @galipremsagarlibraft
binary (#1333) @cjnoletget_worker
(#1365) @pentschevdask
and distributed
for release (#1399) @galipremsagarraft::raft
(#1341) @cjnoletlibraft
binary (#1333) @cjnoletivf_pq
and refine
(#1325) @cjnoletAWS_SESSION_TOKEN
and SCCACHE_S3_USE_SSL
vars to conda build (#1321) @ajschmidt8ivf_pq
serialize/deserialize methods (#1315) @divyegalapydata-sphinx-theme=0.13.0
compatible (#1311) @galipremsagarlibrmm
runtime dependency (#1264) @ajschmidt8dask
and distributed
for development (#1253) @galipremsagarconda
recipe (#1223) @ajschmidt8init_plus_plus
in pylibraft (#1198) @betatimucx-py
version (#1184) @ajschmidt8GenPC
(Permuted Congruential) as the default random number generator everywhere (#1099) @NyrioRAPIDS
magic value (#1132) @robertmaynardeuclidean_dist
in IVF-Flat search (#1122) @Nyrioraft::random::discrete
(#1094) @Nyriodask
and distributed
for release (#1242) @galipremsagarhandle_t
's resources with different workspace_resource (#1165) @cjnoletunary_op
docs and add map_offset
as an improved version of write_only_unary_op
(#1149) @Nyriomatrix::gather
: test coverage, compilation errors, performance (#1126) @Nyrio11.8
and Python 3.10
(#1120) @ajschmidt8Recently Updated
Check (#1117) @ajschmidt8make_blobs
by up to 2x by fixing inefficient kernel launch configuration (#1100) @NyrioGenPC
(Permuted Congruential) as the default random number generator everywhere (#1099) @Nyriodask
and distributed
for development (#1068) @galipremsagarreduce_cols_by_key
to limit atomic conflicts (#1050) @Nyriorandom::discrete
prim (#1039) @Nyriolibraft
binary (#1333) @cjnoletraft::raft
(#1341) @cjnoletlibraft
binary (#1333) @cjnoletivf_pq
and refine
(#1325) @cjnoletAWS_SESSION_TOKEN
and SCCACHE_S3_USE_SSL
vars to conda build (#1321) @ajschmidt8ivf_pq
serialize/deserialize methods (#1315) @divyegalapydata-sphinx-theme=0.13.0
compatible (#1311) @galipremsagarlibrmm
runtime dependency (#1264) @ajschmidt8dask
and distributed
for development (#1253) @galipremsagarconda
recipe (#1223) @ajschmidt8init_plus_plus
in pylibraft (#1198) @betatimucx-py
version (#1184) @ajschmidt8except +
to cython extern cdef declarations (#1001) @benfredmean_squared_error.cuh
(#906) @cjnoletweighted_mean
and add raft::stats
tests (#910) @lowenerraft::stats
API with mdspan (#802) @lowenerdask
and distributed
for release (#1062) @galipremsagarraft
docs theme to pydata-sphinx-theme
(#1026) @galipremsagardask
and distributed
for release" (#1023) @galipremsagardask
and distributed
for release (#1022) @galipremsagardots_along_rows
with rowNorm
and improve coalescedReduction
performance (#1011) @Nyriopylibraft.common.device_ndarray
(#1008) @cjnoletnormalize_rows
in ann_utils.cuh
by a new rowNormalize
prim and improve performance for thin matrices (small n_cols
) (#979) @Nyriolinalg::dot
in public API (#968) @benfredlinalg::mean_squared_error
(#961) @benfredlinalg::axpy
(#942) @benfredraft::matrix
(#937) @lowenerfused_l2_nn_argmin
wrapper to Pylibraft (#924) @cjnoletraft::distance::kernels
(#920) @cjnoletraft::spatial::knn
-> raft::neighbors
(#914) @cjnoletargmin_along_rows
in ANN kmeans (#912) @Nyriomap_along_rows
with matrixVectorOp
(#911) @Nyrioaccumulate_into_selected
from ANN utils into linalg::reduce_rows_by_keys
(#909) @Nyriocub::KeyValuePair
-> raft::KeyValuePair
(#905) @cjnoletdask
and distributed
for development (#886) @galipremsagarexcept +
to cython extern cdef declarations (#1001) @benfredmean_squared_error.cuh
(#906) @cjnoletweighted_mean
and add raft::stats
tests (#910) @lowenerraft::stats
API with mdspan (#802) @lowenerdask
and distributed
for release (#1062) @galipremsagarraft
docs theme to pydata-sphinx-theme
(#1026) @galipremsagardask
and distributed
for release" (#1023) @galipremsagardask
and distributed
for release (#1022) @galipremsagardots_along_rows
with rowNorm
and improve coalescedReduction
performance (#1011) @Nyriopylibraft.common.device_ndarray
(#1008) @cjnoletnormalize_rows
in ann_utils.cuh
by a new rowNormalize
prim and improve performance for thin matrices (small n_cols
) (#979) @Nyriolinalg::dot
in public API (#968) @benfredlinalg::mean_squared_error
(#961) @benfredlinalg::axpy
(#942) @benfredraft::matrix
(#937) @lowenerfused_l2_nn_argmin
wrapper to Pylibraft (#924) @cjnoletraft::distance::kernels
(#920) @cjnoletraft::spatial::knn
-> raft::neighbors
(#914) @cjnoletargmin_along_rows
in ANN kmeans (#912) @Nyriomap_along_rows
with matrixVectorOp
(#911) @Nyrioaccumulate_into_selected
from ANN utils into linalg::reduce_rows_by_keys
(#909) @Nyriocub::KeyValuePair
-> raft::KeyValuePair
(#905) @cjnoletdask
and distributed
for development (#886) @galipremsagar