Romain Lopez

Results 10 comments of Romain Lopez

Ran into a similar error a few years later, it'd be great to have a more precise install procedure :) this thread helped a lot!

Hi @sjfleming, thanks for posting this issue. I agree with Adam, indeed I think it's surprising the early stopping did not converge after much less than 50 iterations. We are...

Remark to self, it would be great to manually check if we can reproduce this bug, as scalability is an important feature of the codebase.

should we create a random anndata matrix of large size (1M per 20k) and try to run scvi overnight?

@khalilouardini would you like to work on this issue? it should be a quick one! Please try to run scVI with either a simulated AnnData or with the 1.3M cells...

yes that sounds like a good advice, it'd be great to put this in the tutorial maybe?

Ran into this today :) it's not fun when analyzing a large data set! Although, agreed that it's a easy fix

I did not get any answers, so I reimplemented stereoscope on another codebase. It takes now a couples of lines of code to run stereoscope from AnnData, and it works...

Also, I found that the number of epochs per default can be a bit conservative. The model can run for many less, and therefore much quicker.

Thanks! You should be able to checkout this branch and install scvi-tools in editable mode; and then you can directly run the notebook (data is not on the branch, but...