predict.cgpca: Predict Convex Generalized PCA scores or reconstruction on...

In andland/generalizedPCA: Generalized PCA

Description

Predict Convex Generalized PCA scores or reconstruction on new data

Usage

## S3 method for class 'cgpca'
predict(object, newdata, type = c("PCs", "link", "response"),
  ...)

Arguments

object	convex generalized PCA object
newdata	matrix with all binary entries. If missing, will use the data that object was fit on
type	the type of fitting required. type = "PCs" gives the PC scores, type = "link" gives matrix on the logit scale and type = "response" gives matrix on the probability scale
...	Additional arguments

Examples

# construct a low rank matrices in the logit scale
rows = 100
cols = 10
set.seed(1)
loadings = rnorm(cols)
mat_logit = outer(rnorm(rows), loadings)
mat_logit_new = outer(rnorm(rows), loadings)

# convert to a binary matrix
mat = (matrix(runif(rows * cols), rows, cols) <= inv.logit.mat(mat_logit)) * 1.0
mat_new = (matrix(runif(rows * cols), rows, cols) <= inv.logit.mat(mat_logit_new)) * 1.0

# run generalized PCA on it
cgpca = convexGeneralizedPCA(mat, k = 1, M = 4, family = "binomial")

PCs = predict(cgpca, mat_new)

andland/generalizedPCA documentation built on May 12, 2019, 2:42 a.m.