Skip to main content
. 2020 Jan 8;36(8):2401–2409. doi: 10.1093/bioinformatics/btaa003

Fig. 2.

Fig. 2.

Dependence of the EC classification accuracy (Level 1; EC50 dataset) on the size of the training dataset. UDSMProt performs better than the baseline model also in the regime of small datasets that is particularly important for practical applications