-
When running
`julia --code-coverage=all --inline=no`
```
import(CoverageBase)
using Base.Test
CoverageBase.runtests(CoverageBase.testnames())
```
On OSX 10.10.3, and on the Coveralls builds on Ubunt…
-
For small complex eigenvalues of the Hessenberg matrix, `beast.evolution.substitutionmodel.DefaultEigenSystem.hqr2` tries to access index `-1` of `double[][] h`.
When `en==1`, the loop starting in li…
-
As discussed on the mailing list, it would be good to have the option to specify the rank of the noise covariance matrix when computing an inverse operator, like in the MNE command line tools.
After …
-
**Reported by marcus on 20 Apr 43350065 14:46 UTC**
Hello,
I've tested the Kernel PCA code and got some unexpected results.
So I looked into the code (kernel_pca_impl.hpp). I suppose it's not the ri…
-
When trying to construct a quadratic function from a vector variable and a CVXOPT sparse matrix, I receive a ValueError.
Simplified code:
```
import cvxpy
import cvxopt
Q = cvxopt.spdiag([1,1])
x = …
-
This entrance task is about polishing Shogun's PCA implementation, which currently works, but may be slow and is not tested.
- Eigen3-ise everything, currently its in lapack. See for example PR #1874 …
-
When I use EJMLUtils.class to do eigendecomposition of the rate matrix when the dimension of the rate matrix is 20_20 but it is represented by 36 features, during mcmc iterations, we might encounter t…
-
From #42.
-
It appears that sometimes the sign of `X` is switched compared to `Y` when embedding a directed graph. For example, we should have below that the mean is around .3 for the outer product between X and …
-
This might not be bug in scs but a problem due to using prebuilt blas and lapack libraries. I am using windows 8.1 and anaconda. I downloaded the prebuilt 64bit libs and dlls for blas and lapack from …
ncroc updated
10 years ago