explore setting some training data aside for final validation of released models

openvax / mhcflurry

Peptide-MHC I binding affinity prediction

http://openvax.github.io/mhcflurry/

Apache License 2.0

193 stars 58 forks source link

explore setting some training data aside for final validation of released models #56

Closed timodonnell closed 6 years ago

timodonnell commented 8 years ago

Currently we have no scores to report for the released models since we have used all our data to train them. Setting aside a small amount of data (perhaps 50 points per allele?) would enable testing the accuracy of the released models, of course at the cost of having less training data.