compomics / DeepLC

DeepLC: Retention time prediction for (modified) peptides using Deep Learning.
https://iomics.ugent.be/deeplc
Apache License 2.0
56 stars 19 forks source link

Optimized library usage (~85% reduce in DeepLC prediction time) #49

Closed markmipt closed 1 year ago

markmipt commented 2 years ago

Hi everyone!

I've optimized code when library is used for RT prediction. In my case, when library contains 6 million lines, I have 80-90% reduction in time for the full DeepLC prediction workflow for both small (~10k) and large sets (~750k) of peptides in prediction.

The most important part is just deleting of "idents_in_lib = set(LIBRARY.keys())" line. However, the rest of the code optimization could be also important in some specific cases.

Regards, Mark