Arta Seyedian

> Try to add this line in models.py > > torch.backends.cudnn.enabled=False This worked, but now I'm running into this issue: ``` ======================================= Start to generate images Traceback (most recent call...

@DanBigioi I don't remember. I did have a typo though; I am using a 3080 and not a 2080. I think I settled on not having enough memory.

> Have you tried run integrate module instead of breaking it into 3 steps? No, I haven't. I'll give it a shot, might help. > or may need to increase...

I tried to use the `integrate` module instead of breaking it up by the submodules like before: ``` # Array of subdirectory paths SUBDIRS=( "day_3_scATACpro/ALL4364_Combo_day_3" "day_3_scATACpro/ALL4364_Control_day_3" "day_3_scATACpro/ALL4364_Rux_day_3" "day_3_scATACpro/ALL4364_Ven_day_3" "day_3_scATACpro/JH331_Combo_day_3" "day_3_scATACpro/JH331_Control_day_3"...

I have this line in the `configure_user.txt`: `Top_Variable_Features = .1 ## number/fraction of variable features used for seurat`

I'll give it a shot and let you know.

I actually haven't had a chance to try it out yet, I'll do it today, thanks for the reminder!

I submitted the batch job just now. One thing I was wondering is if the `integrate` module calls `merge_peaks` and `reConstMtx` as well? I need those submodules for my samples.

Here's the `configure` file: ``` (base) [seyediana@reslnvhpc071 scATAC-pro_singularity]$ grep "Top" configure/integrated/configure_allPeaks_integrated.txt ## Top_Variable_Features = 0.1 ## number/fraction of variable features used for seurat Top_Variable_Features = .2 ## number/fraction of variable...

Just realized this myself. It kind of makes sense since the Obsidian note is the repository of all Anki cards, but is still frustrating to realize when you sync and...