AndreasGerken
AndreasGerken
I thought about the issue for some time aswell. The solution of MrSpiffyClean works fine but is not as simple for the average user as cookies would be.
I have the same issue which makes the package unusable for me sadly :/ I did a optimization for n_components and just at the optimum (~200 components for 6d data)...
Just found the solution. It was already answered in the issues but it wasn't documented. The issue has to do with a too small covariance_regularization. It can be increased to...
Added figure to show issues with overestimation of kl divergence when sample sizes are different.
Would love to use a higher version aswell.
I have the same issue. It appears when using cpu or cuda. Seems like it appears with high n_components or n_features in comparison to samples. Is there a possibility to...
Thanks @larsbratholm. I'm struggling with a complex optimization for my PhD since some month and found your post yesterday. It finally works and you helped me to get the optimization...