Utilities
Batching
RandomFeatures.Utilities.batch_generator
— Functionbatch_generator(
array::AbstractArray,
batch_size::Int64;
dims
) -> Any
produces batched sub-array views of size batch_size
along dimension dims
.
this creates views not copies. Modifying a batch will modify the original!
Matrix Decomposition
RandomFeatures.Utilities.Decomposition
— Typestruct Decomposition{T, M<:(AbstractMatrix), MorF<:Union{AbstractMatrix, LinearAlgebra.Factorization}}
Stores a matrix along with a decomposition T=Factor
, or pseudoinverse T=PseInv
, and also computes the inverse of the Factored matrix (for several predictions this is actually the most computationally efficient action)
full_matrix::AbstractMatrix
: The original matrixdecomposition::Union{AbstractMatrix, LinearAlgebra.Factorization}
: The matrix decomposition, or pseudoinverseinv_decomposition::AbstractMatrix
: The matrix decomposition of the inverse, or pseudoinverse
RandomFeatures.Utilities.StoredInvType
— Typeabstract type StoredInvType
Type used as a flag for the stored Decomposition type
RandomFeatures.Utilities.Factor
— Typeabstract type Factor <: StoredInvType
RandomFeatures.Utilities.PseInv
— Typeabstract type PseInv <: StoredInvType
RandomFeatures.Utilities.get_decomposition
— Functionget_decomposition(
d::Decomposition
) -> Union{AbstractMatrix, LinearAlgebra.Factorization}
get decomposition
field
RandomFeatures.Utilities.get_full_matrix
— Functionget_full_matrix(d::Decomposition) -> AbstractMatrix
get full_matrix
field
RandomFeatures.Utilities.get_parametric_type
— Functionget_parametric_type(
d::Decomposition{T, M<:Union{AbstractMatrix, LinearAlgebra.Factorization}}
) -> Any
get the parametric type
RandomFeatures.Utilities.linear_solve
— Functionlinear_solve(
d::Decomposition,
rhs::AbstractArray{<:AbstractFloat, 3},
::Type{Factor};
tullio_threading
) -> Any
Solve the linear system based on Decomposition
type