Matthew Schmitz
Matthew Schmitz
Yes, I'm also wondering how you get the coefficients for multiple factor levels of a covariate? If you have 4 clusters, how do you get the beta for each of...
I figured out that if you set `pynndescent_n_neighbors` to a lower number it solves this issue. Perhaps there should be an internal conditional or this!
Not intentional, but was working on a subsample of a huge dataset so it makes sense that a few of the batches ended up very small! On Thu, Sep 30,...
FYI: I also took a stab at fixing the straightthroughcategorical, this could still use some work but it works for me where the previous RelaxedCategoricalStraightThrough would not train as part...
Hi @fritzo, I agree in principle, you're right about the the hackability as well as using the proper epsilon or torch tiny, still working on my coding modularity :). I...
@flying-sheep Yeah I can try to make a little notebook, I'm also working on a function to overwrite selected fields as well that I can include. Should I just commit...
Wow I wasn't aware of cap-anndata, that seems like a much more robust solution, I think that should probably be an example notebook rather than my hack solution! On Wed,...
I love it when you start a conversation with an ad hoc approach and end up with several robust, purpose-built solutions :)
I'm also running into this bug :)