ctallec

You are right, we are passing the mean instead of a sample. I don't think this will make a significant difference, and notably it's unclear wether it is going to...

@wildermuthn You may want to be extra cautious with the hyperparameters you use and the training duration, typically you want to use the same hyperparameters for cma as in the...

Hi @MilesCranmer, Seems like a problem with the specific implementation of LSTMs, I will look into that asap and keep you updated. Thanks for reporting the issue :+1: