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.

Linear Algebra

This page provides C++ class references for the publicly-exposed elements of the raft/linalg (dense) linear algebra headers. In addition to providing highly optimized arithmetic and matrix/vector operations, RAFT provides a consistent user experience by providing common BLAS routines, standard linear system solvers, factorization and eigenvalue solvers. Some of these routines hide the complexities of lower-level C-based libraries provided in the CUDA toolkit

Contents:

• Arithmetic
  • Addition
  • Binary Op
  • Division
  • Multiplication
  • Power
  • Square Root
  • Subtraction
  • Ternary Op
  • Unary Op
• BLAS Routines
  • axpy
  • dot
  • gemm
  • gemv
• Mapping and Reduction
  • Coalesced Reduction
  • Map
  • Map Reduce
  • Mean Squared Error
  • Norm
  • Normalize
  • Reduction
  • Reduce Cols By Key
  • Reduce Rows By Key
  • Strided Reduction
• Matrix Operations
  • Transpose
• Matrix-Vector Operations
  • Arithmetic
  • Operations
• Linear Algebra Solvers
  • Eigen Decomposition
  • QR Decomposition
  • Randomized Singular-Value Decomposition
  • Singular-Value Decomposition
  • Least Squares