Karin Hrovatin
Karin Hrovatin
I think you can add 10x reading function from Compass, they seem to have similar inputs https://github.com/YosefLab/Compass/blob/abc4456e0b4752b581b3fdfa68882d72e60ab307/compass/utils.py#L56
Also, can you note whether the input should be raw counts, normalised counts, or normalised+log transformed counts.
Does this mean I can also input log-transformed normalised raw counts?
Here is my solution. Please note some TODOs as I think that how the PDB is currently parsed may be problematic - would recomend using BioPython instead. - I tried...
Basically using a PDB file with two chains (sorry cant share ours) and setting `plddt_mask=True`
This is not ideal as then it may be unclear if utils should be loaded from SaProt or my own utils directory that is somewhere in the path Best solution...
Does anyone have a code snippet showing how the issue with LKJCovariance prior can be fixed? EDIT: Example for closure of `LKJCovariancePrior` on `IndexKernel` here: https://github.com/pytorch/botorch/discussions/2656#discussioncomment-11785034
Thank you very much for the response. I think wasn't clear - I meant that using `StratifiedStandardize` would prevent using the multi-output option of the `MultiTaskGP`, if I am not...
@Scienfitz I would keep open as the main blocker for this was randomness in benchmarks. Since that may be solved now I would suggest running benchmarks again on the new...
No, I need to first set up testing on oneHPC to reproducibly benchmark - as that seems to be the only option to make fully reproducible. I will post update...