Shams

Seems to be something to do with the matrix dimensions as @Hrovatin. Test breaks until 129 features and works smoothly after that with the simulated data as in the tutorial...

Sorry for the confusion. I wanted to ask the extend of your suggestion. I want to ask like how many cells have activation of ``` gene set=gene1,2....n ``` at a...

could resolve by giving the `conservation_file` path in the command line as worked in my example ``` ./vep --plugin LoF,loftee_path:/Tools/ensembl-vep/loftee,human_ancestor_fa:/SupportFiles/vep/LOFTEE/human_ancestor.fa.gz,phylocsf_data:/SupportFiles/vep/LOFTEE/phylocsf_gerp.sql ``` . Hope this help!

Getting similar error. Tried increasing the plot size and still have the same error ``` Error in .local(GdObject, ...): Too many stacks to draw. Either increase the device size or...

@ccruizm seems you can invoke the GPU utility by ``` from spectra import spectra_gpu as spc_gpu ``` and all the commands are available with `spc_gpu.'...'`. It seems still experimental and...

That’s a great improvement but not sure why it’s not showing in the resource usage. Is there any branch that I can try the new implementation?.

Thank you for very quick response. Are there any specific commands or preprocesses required? I tried it yesterday. CPU version is running fine but when I tried on GPU version,...

Yes, I have GPU detected in the session. Just modified the snippets above ``` import torch # Check if CUDA is available if torch.cuda.is_available(): print("CUDA is available") # Get the...

@russellkune Seems there are two independent problems. 1. The delay I was observing due to the issue of running diffrent GPU jobs on the same GPU core. I was running...

Finally it works (n#1 did work after about an hour stalling. Not sure what is causing it, but once it started the process, it was much faster, yay!). Are there...