Incomplete: create a matrix of class 'Incomplete'
In softImpute: Matrix Completion via Iterative Soft-Thresholded SVD

Description Usage Arguments Details Value Author(s) See Also Examples

creates an object of class Incomplete, which inherits from class dgCMatrix, a specific instance of class sparseMatrix

1	Incomplete(i, j, x)

`i`	row indices
`j`	column indices
`x`	a vector of values

The matrix is represented in sparse-matrix format, except the "zeros" represent missing values. Real zeros are represented explicitly as values.

a matrix of class Incomplete which inherits from class dgCMatrix

Trevor Hastie and Rahul Mazumder

softImpute

set.seed(101)
n=200
p=100
J=50
np=n*p
missfrac=0.3
x=matrix(rnorm(n*J),n,J)%*%matrix(rnorm(J*p),J,p)+matrix(rnorm(np),n,p)/5
ix=seq(np)
imiss=sample(ix,np*missfrac,replace=FALSE)
xna=x
xna[imiss]=NA
xnaC=as(xna,"Incomplete")
### here we do it a different way to demonstrate Incomplete
### In practise the observed values are stored in this market-matrix format.
i = row(xna)[-imiss]
j = col(xna)[-imiss]
xnaC=Incomplete(i,j,x=x[-imiss])