Just checked, QR is exposed by netlib: import org.netlib.lapack.Dgeqrf
For the equality and bound version, I will use QR...it will be faster than
the LU that I am using through jblas.solveSymmetric...
On Thu, Oct 16, 2014 at 8:34 AM, Debasish Das
wrote:
> @xiangrui should we add this epsilon in
@xiangrui should we add this epsilon inside ALS code itself ? So that if
user by mistake put 0.0 as regularization, LAPACK failures does not show
up...
@sean For the proximal algorithms I am using Cholesky for L1 and LU for
equality and bound constraints (since the matrix is quasi definite)...I am
It Gramian is at least positive semidefinite and will be definite if the
matrix is non singular, yes. That's usually but not always true.
The lambda*I matrix is positive definite, well, when lambda is positive.
Adding that makes it definite.
At least, lambda=0 could be rejected as invalid.
But t
Do not use lambda=0.0. Use a small number instead. Cholesky
factorization doesn't work on semi-positive systems with 0
eigenvalues. -Xiangrui
On Wed, Oct 15, 2014 at 5:05 PM, Debasish Das wrote:
> But do you expect the mllib code to fail if I run with 0.0 regularization ?
>
> I think ||r - wi'hj|
But do you expect the mllib code to fail if I run with 0.0 regularization ?
I think ||r - wi'hj||^{2} is positive definite...It can become positive
semi definite only if there are dependent rows in the matrix...
@sean is that right ? We had this discussion before as well...
On Wed, Oct 15, 2014
Hi Debaish,
I think ||r - wi'hj||^{2} is semi-positive definite.
Thanks,
Liquan
On Wed, Oct 15, 2014 at 4:57 PM, Debasish Das
wrote:
> Hi,
>
> If I take the Movielens data and run the default ALS with regularization as
> 0.0, I am hitting exception from LAPACK that the gram matrix is not
> pos
