Attention

The vector search and clustering algorithms in RAFT are being migrated to a new library dedicated to vector search called cuVS. We will continue to support the vector search algorithms in RAFT during this move, but will no longer update them after the RAPIDS 24.06 (June) release. We plan to complete the migration by RAPIDS 24.10 (October) release and they will be removed from RAFT altogether in the 24.12 (December) release.

BLAS Routines#

axpy#

#include <raft/linalg/axpy.cuh>

namespace raft::linalg

template<typename ElementType, typename IndexType, typename InLayoutPolicy, typename OutLayoutPolicy, typename ScalarIdxType>
void axpy(raft::resources const &handle, raft::device_scalar_view<const ElementType, ScalarIdxType> alpha, raft::device_vector_view<const ElementType, IndexType, InLayoutPolicy> x, raft::device_vector_view<ElementType, IndexType, OutLayoutPolicy> y)#

axpy function It computes the following equation: y = alpha * x + y

Parameters:

  • handle – [in] raft::resources

  • alpha – [in] raft::device_scalar_view

  • x – [in] Input vector

  • y – [inout] Output vector

template<typename ElementType, typename IndexType, typename InLayoutPolicy, typename OutLayoutPolicy, typename ScalarIdxType>
void axpy(raft::resources const &handle, raft::host_scalar_view<const ElementType, ScalarIdxType> alpha, raft::device_vector_view<const ElementType, IndexType, InLayoutPolicy> x, raft::device_vector_view<ElementType, IndexType, OutLayoutPolicy> y)#

axpy function It computes the following equation: y = alpha * x + y

Parameters:

  • handle – [in] raft::resources

  • alpha – [in] raft::device_scalar_view

  • x – [in] Input vector

  • y – [inout] Output vector

dot#

#include <raft/linalg/dot.cuh>

namespace raft::linalg

template<typename ElementType, typename IndexType, typename ScalarIndexType, typename LayoutPolicy1, typename LayoutPolicy2>
void dot(raft::resources const &handle, raft::device_vector_view<const ElementType, IndexType, LayoutPolicy1> x, raft::device_vector_view<const ElementType, IndexType, LayoutPolicy2> y, raft::device_scalar_view<ElementType, ScalarIndexType> out)#

Computes the dot product of two vectors.

Parameters:

  • handle – [in] raft::resources

  • x – [in] First input vector

  • y – [in] Second input vector

  • out – [out] The output dot product between the x and y vectors.

template<typename ElementType, typename IndexType, typename ScalarIndexType, typename LayoutPolicy1, typename LayoutPolicy2>
void dot(raft::resources const &handle, raft::device_vector_view<const ElementType, IndexType, LayoutPolicy1> x, raft::device_vector_view<const ElementType, IndexType, LayoutPolicy2> y, raft::host_scalar_view<ElementType, ScalarIndexType> out)#

Computes the dot product of two vectors.

Parameters:

  • handle – [in] raft::resources

  • x – [in] First input vector

  • y – [in] Second input vector

  • out – [out] The output dot product between the x and y vectors.

gemm#

#include <raft/linalg/gemm.hpp>

namespace raft::linalg

template<typename ValueType, typename IndexType, typename LayoutPolicyX, typename LayoutPolicyY, typename LayoutPolicyZ, typename ScalarIdxType = std::uint32_t, typename ScalarViewType = raft::host_scalar_view<ValueType, ScalarIdxType>, typename = std::enable_if_t<std::disjunction_v<std::is_same<ScalarViewType, raft::host_scalar_view<ValueType, ScalarIdxType>>, std::is_same<ScalarViewType, raft::device_scalar_view<ValueType, ScalarIdxType>>>>>
void gemm(raft::resources const &res, raft::device_matrix_view<ValueType, IndexType, LayoutPolicyX> x, raft::device_matrix_view<ValueType, IndexType, LayoutPolicyY> y, raft::device_matrix_view<ValueType, IndexType, LayoutPolicyZ> z, std::optional<ScalarViewType> alpha = std::nullopt, std::optional<ScalarViewType> beta = std::nullopt)#

GEMM function designed for handling all possible combinations of operand layouts (raft::row_major or raft::col_major) with scalars alpha and beta on the host or device It computes the following equation: Z = alpha . X * Y + beta . Z If alpha is not provided, it is assumed to be 1.0 If beta is not provided, it is assumed to be 0.0.

Template Parameters:

  • ValueType – Data type of input/output matrices (float/double)

  • IndexType – Type of index

  • LayoutPolicyX – layout of X

  • LayoutPolicyY – layout of Y

  • LayoutPolicyZ – layout of Z

Parameters:

gemv#

#include <raft/linalg/gemv.cuh>

namespace raft::linalg

template<typename ValueType, typename IndexType, typename LayoutPolicy, typename ScalarIdxType = std::uint32_t, typename ScalarViewType = raft::host_scalar_view<ValueType, ScalarIdxType>, typename = std::enable_if_t<std::disjunction_v<std::is_same<ScalarViewType, raft::host_scalar_view<ValueType, ScalarIdxType>>, std::is_same<ScalarViewType, raft::device_scalar_view<ValueType, ScalarIdxType>>>>>
void gemv(raft::resources const &handle, raft::device_matrix_view<const ValueType, IndexType, LayoutPolicy> A, raft::device_vector_view<const ValueType, IndexType> x, raft::device_vector_view<ValueType, IndexType> y, std::optional<ScalarViewType> alpha = std::nullopt, std::optional<ScalarViewType> beta = std::nullopt)#

GEMV function designed for raft::col_major layout for A It computes y = alpha * op(A) * x + beta * y, where length of y is number of rows in A while length of x is number of columns in A If layout for A is provided as raft::row_major, then a transpose of A is used in the computation, where length of y is number of columns in A while length of x is number of rows in A If alpha is not provided, it is assumed to be 1.0 If beta is not provided, it is assumed to be 0.0.

Template Parameters:

  • ValueType – Data type of input/output matrices (float/double)

  • IndexType – Type of index

  • LayoutPolicyX – layout of X

  • LayoutPolicyY – layout of Y

  • LayoutPolicyZ – layout of Z

Parameters: