For small particle sets, like at the end of the data processing journey, a modest fraction of micrographs are no longer contributing. Extractions and RBMC and Topaz require micrograph inputs for the particle set. Instead of adding unneeded micrographs, it’s possible to run a preparative job to curate the micrographs by supplying particles and setting cutoff to >0 for particles per micro. Would it be possible (easy) to have these and other job types perform this curation in job preprocessing?
use case: I provide 20,000 particles for extraction from 10,000 micrographs, job quickly calculates that only 6,200 micrographs are needed, auto-skips 3,800 (and potentially provides an output of unused micrographs).
it seems like extracting (or training Topaz picking or RBMC hyperparameters etc) takes an equal amount of time to realize the output is zero particles for unused micrographs.