1

I am trying to use the cholesky decomposition to invert a sparse matrix and I found this great package to help me on my problem. My issue is that I keep getting CholmodNotPositiveDefiniteError although I verified with the eigenvalues that my matrix is positive and I tried adding small positive diagonal matrix but it didn't work. Here's my notebook

Could you please help me on this?

enter image description here

Zephyr
  • 11,891
  • 53
  • 45
  • 80
mirimo
  • 144
  • 4
  • Just in case: are you sure that the matrix is symmetric? In numpy, `X * Y` means element-wise product, not matrix product; I suspect that it's not what intedned. –  Jul 08 '20 at 09:42
  • @Dmitry I don't think that's the case for sparse matrix , I've verified that it's equivalent to the X.T.dot(X) – mirimo Jul 08 '20 at 09:54

0 Answers0