Hi there
I'm analysing one single cell RNA seq dataset with PYSCENIC (useful for transcription factors). The first network inference step is “pyscenic grn”, and I’ve decided to use the grnboost2 inference algorithm from the CLI because of the dimensions of my dataset (25k cells).
However, I’m getting a lot of warnings (“full garbage collection took 10% CPU time recently (threshold: 10%)”, “Worker exceeded 95% memory budget. Restarting”) and some errors.
Could any of you help me? Did you get the same problems using Pyscenic with “large” datasets? How did you solve them?
Please note: before trying on my dataset, I used the “tutorial” dataset. I got the same warnings but finally it worked. While actually with my “real” dataset the function has been running for 15 hours at least and at the moment it has not finished yet.
