IExtendedLinalgBackend
Interfaces · AiDotNet.Tensors.Engines.DirectGpu
No summary documentation available yet.
Known implementations: CudaBackend, HipBackend, MetalBackend, OpenClBackend, VulkanBackend, WebGpuBackend
Methods
| Method | Summary |
|---|---|
void LinalgCholeskyBackwardLower(IGpuBuffer factor, IGpuBuffer gradOutput, IGpuBuffer gradInput, int batchCount, int n) |
|
void LinalgEighSymmetric(IGpuBuffer input, IGpuBuffer eigenvalues, IGpuBuffer eigenvectors, int batchCount, int n, bool upper) |
|
void LinalgLdlFactor(IGpuBuffer input, IGpuBuffer ld, IGpuBuffer pivots, int batchCount, int n, bool upper) |
|
void LinalgLdlSolve(IGpuBuffer ld, IGpuBuffer pivots, IGpuBuffer rhs, IGpuBuffer solution, int batchCount, int n, bool upper, bool rhsIsVector) |
|
void LinalgLuSolve(IGpuBuffer lu, IGpuBuffer pivots, IGpuBuffer rhs, IGpuBuffer solution, int batchCount, int n, int rightHandSides, bool rhsIsVector) |
|
void LinalgQrReducedDirect(IGpuBuffer input, IGpuBuffer q, IGpuBuffer r, int batchCount, int m, int n) |
|
void LinalgSolveBackwardVector(IGpuBuffer input, IGpuBuffer solution, IGpuBuffer gradOutput, IGpuBuffer gradInput, IGpuBuffer gradRhs, int batchCount, int n) |
|
void LinalgSolveVector(IGpuBuffer input, IGpuBuffer rhs, IGpuBuffer solution, IGpuBuffer info, int batchCount, int n) |
|
void LinalgSvdReduced(IGpuBuffer input, IGpuBuffer u, IGpuBuffer singularValues, IGpuBuffer vh, int batchCount, int m, int n) |
|
void LinalgTriangularSolveVector(IGpuBuffer input, IGpuBuffer rhs, IGpuBuffer solution, int batchCount, int n, bool upper, bool unitDiagonal) |