Rinalmo for sequence prediction

#1
by NaimaVahab - opened

I tried to implement splice site prediction for acceptor and donor using the sequence prediction head implemented here. But it is giving very low metrics ; such as F1 of 0.33 only whereas actual paper reports F1 around 98

MultiMolecule org

Thank you for your feedback.
We have confirmed that the model code and weights provided in this repository matches the version released by the author of RiNALMo, and the model yields identical output as the version released by the author of RiNALMo.
Please contact the authors of RiNALMo for any issues in reproduce.

Sign up or log in to comment