Hey all, I hope all is well. I am a little confused with the behavior of SparseDoubleMatrix2D. I am using .9.4 and am calling Colt from the most recent version of scala.
When I call
val m = new SparseDoubleMatrix2D(100,100,Array.range(0,100),Array.range(0,100),1)
and then
val n = m.zMult(m,null,2,0,false,false)
I would expect that n would be a sparse matrix. Instead, it is dense with storage of all of the zeros. I have
tried the same thing through the factory method creation and again,
Thanks in advance!
m
Actually, I am seven more confused.
val m = DoubleFactory2D.sparse
n = m.make(4,4)
// Set n to be identity
Creating a DoubeProperty d then throws an exception when called with d.checksparse( n)
Printing out n at the REPL gives
n
res27: cern.colt.matrix.tdouble.DoubleMatrix2D =
4 x 4 sparse matrix, nnz = 4
(0,0) 1.0
(1,1) 1.0
(2,2) 1.0
(3,3) 1.0
Thanks!
Interestingly, checkDense(n) also throws an exception saying the matrix is not dense.. ALso, if I take n and cast it to a Sparse Matrix, I still get the exception for checkSparse()
More debugging, it see if I make the sparse matrix a row compressed matrix then things go through...