trunc.svd: Truncated singular value decomposition
In lfa: Logistic Factor Analysis for Categorical Data

Description Usage Arguments Details Value

Truncated SVD

1
2
3

## S3 method for class 'svd'
trunc(A, d, adjust = 3, tol = 1e-10, V = NULL,
  seed = NULL, ltrace = FALSE, override = FALSE)

`A`	matrix
`d`	number of singular vectors
`adjust`	extra singular vectors to calculate for accuracy
`tol`	convergence criterion
`V`	optional initial guess
`seed`	seed
`ltrace`	debugging output
`override`	TRUE means we use fast.svd instead of the iterative algorithm (useful for small data or very high d).

Performs singular value decomposition but only returns the first d singular vectors/values. The truncated SVD utilizes Lanczos bidiagonalization. See references.

This function was modified from the package irlba 1.0.1 under GPL. Replacing the crossprod() calls with the C wrapper to dgemv is a dramatic difference in larger datasets. Since the wrapper is technically not a matrix multiplication function, it seemed wise to make a copy of the function.

list with singular value decomposition.

lfa documentation built on Nov. 8, 2020, 7:30 p.m.