Matthew Schmitz

Maybe add epsilon to RelaxedOneHotCategorical to prevent underflow

3

I've noticed that pyro.distributions.RelaxedOneHotCategorical tends to underflow pretty dramatically if you decrease the temperature below 0.3 or so with many categories. I've been adding a slight modification to the rsample...

ValueError: No hyperplanes of adequate size were found! When not using annoy

5

Hi there, Having an issue when I try to run BBKNN without annoy. Had this error, then freshly installed everything in a new conda environment, I'm still getting the error...

Torchscript error in JitTraceEnum_ELBO Torch Version 2.2.1, CUDA Version: 12.3

1

Hi there, Noticed a bug in JitTraceEnum_ELBO. My code runs fine with a previous version of pytorch or with JitTrace_ELBO (I can use RelaxedOneHotCategorical instead of OneHotCategorical for what I...

Reading Anndata from only parts of h5ad file: Hack solution

8

I came across a previous issue #436 and couldn't get the dask solution working with my application, so I came up with a somewhat hacky solution to reading only the...

Hi there, When I run code like this: ``` data = snap.read_dataset("concatenated.h5ads",mode='r+') data.obs['doubcall']=list(np.random.choice([True,False],data.shape[0])) data.subset(obs_indices=pd.Series(data.obs_names)[~data.obs['doubcall']],out="concatenated_nodoublet") ``` It creates a directory concatenated_nodoublet, but this contains no files, while all of the h5ads...

TypeError when writing string columns to h5ad

2

### Please make sure these conditions are met - [X] I have checked that this issue has not already been reported. - [X] I have confirmed this bug exists on...