The pseudoinverse may be defined algebraically_
but it is more computationally convenient to understand it through the SVD_
Supports input of float, double, cfloat and cdouble dtypes.
Also supports batches of matrices, and if A is a batch of matrices then
the output has the same batch dimensions.
linalg_pinv(A, rcond = NULL, hermitian = FALSE, atol = NULL, rtol = NULL)(Tensor): tensor of shape (*, m, n) where * is zero or more batch dimensions.
(float or Tensor, optional): the tolerance value to determine when is a singular value zero
If it is a torch_Tensor, its shape must be
broadcastable to that of the singular values of
A as returned by linalg_svd().
Alias for rtol.
Default: 0.
(bool, optional): indicates whether A is Hermitian if complex
or symmetric if real. Default: FALSE.
the absolute tolerance value. When NULL it’s considered to be zero.
the relative tolerance value. See above for the value it takes when NULL.
If hermitian= TRUE, A is assumed to be Hermitian if complex or
symmetric if real, but this is not checked internally. Instead, just the lower
triangular part of the matrix is used in the computations.
The singular values (or the norm of the eigenvalues when hermitian= TRUE)
that are below the specified rcond threshold are treated as zero and discarded
in the computation.
linalg_inv() computes the inverse of a square matrix.
linalg_lstsq() computes A$pinv() %*% B with a
numerically stable algorithm.
Other linalg:
linalg_cholesky_ex(),
linalg_cholesky(),
linalg_det(),
linalg_eigh(),
linalg_eigvalsh(),
linalg_eigvals(),
linalg_eig(),
linalg_householder_product(),
linalg_inv_ex(),
linalg_inv(),
linalg_lstsq(),
linalg_matrix_norm(),
linalg_matrix_power(),
linalg_matrix_rank(),
linalg_multi_dot(),
linalg_norm(),
linalg_qr(),
linalg_slogdet(),
linalg_solve(),
linalg_svdvals(),
linalg_svd(),
linalg_tensorinv(),
linalg_tensorsolve(),
linalg_vector_norm()
if (torch_is_installed()) {
A <- torch_randn(3, 5)
linalg_pinv(A)
}
Run the code above in your browser using DataLab