stereoscope

stereoscope copied to clipboard

test error

1

Hi, Thanks for sharing the code. Following the guidance, my installation of stsc on a Linux server was successful. Running `python3 -c "import stsc; print(stsc.__version__)"` also gave me output as...

linhuawang

The `--sc_model` option hints that it might be possible to 'recycle' an sc model that has already been fitted in a previous `run`. Is this the case? If so, how?...

johnyaku

Strategy for selections of genes

4

Hi Alma, The current strategy (if I am not mistaken) is to fit the models from the SC data using either all or top N genes. When the top N...

jfnavarro

Number of cores/threads

1

Hi, is there a way to limit stereoscope to a certain number of threads/cores? I am running it on a server where other users need to work too, so I...

kdragicevic

counts of gene g at spot s from cell c?

2

Hi! Thank you for developing such a nice tool! I guess the fit_sc_data function returns the parameters that are obtained from SC data's negative binomial distribution. Is this the same...

gorkemkose

Top n criteria and some other things

1

* Add option to select the top n criteria (expression or variance) * Fix a bug where the top N selection of genes was being applied to the ST data...

jfnavarro

Hi, thank you for the nice work! I find that when handling both mouse brain ST / Visium data and developmental heart spatial data, you excluded all spots that were...

jiazhao97

Interaction with AnnData

6

I am trying to work outside of python command line and more in interaction with AnnData / scanpy. I am tempted to fork this package and patch it to make...

romain-lopez

--silent_mode still prints output

2

Depsite the flag set, output of training is still printed. Would be cool to just return single value at end of training.

giovp

Hi, so I happen to have really low gpu utilization with default parameters (~10%), if I increase batch size to 500 I get to 20%, in both cases on a...

giovp