redst4r

keeps happening to me too in kallisto 0.46.2: ``` [quant] finding pseudoalignments for the reads ... [quant] done [quant] processed 250,960,675 reads, 156,761,018 reads pseudoaligned [ em] quantifying the abundances...

just changing the existing `tf.XYZ()` calls into the corresponding `keras.backend.` calls doesnt work?

Hi, thanks, I'll try editing the dataset.json (unfortunately, I got rid of the intermediate files created by the scanpy-export, so I cant quickly rebuild the html). An input box with...

Just had a look at the `dataset.json` file in the html directory. Here's how it looks like, I don't see a "radius" setting, not sure exactly where to add it:...

great, editing the `dataset.json` of the actual dataset worked nicely as a quick fix!! In the future, I'll make sure radius is already set when building the cellbrowser-website!

same here

Hi, I have a similar problem, i.e. I want to apply the same set of convolutions (shared weights) to lots of different 'color' channels. As suggested by @f0k, I slice...

same problem here. Not sure if this is related, but opening the `Options`, I get a ``` Error handling response: TypeError: options.feedList.join is not a function at TMNSetOptionsMenu (chrome-extension://cgllkjmdafllcidaehjejjhpfkmanmka/option-script.js:152:37) at...

Same issue as @Avaptel18 on writing to hdf5: ``` TypeError: Can't implicitly convert non-string objects to strings ``` It happens if you calculate marker genes with `.rank_genes_groups()` and filter them...

Here's a minimal example of the bug I run into: ```python import scanpy as sc import anndata print("scanpy", sc.__version__) # 1.9.3 print("anndata", anndata.__version__) # 0.8.0 adata = sc.datasets.pbmc3k_processed() sc.tl.rank_genes_groups(adata, groupby='louvain')...