ESM-2 for Binding Site Prediction

This model is overfit.

Training procedure

Epoch: 2
Training Loss: 0.051500
Validation Loss: 0.417758
Accuracy: 0.951942
Precision: 0.422039
Recall: 0.747092
F1 Score: 0.539379
AUC: 0.853526
MCC: 0.539618

Framework versions

PEFT 0.5.0

Downloads last month: 21

Inference Examples

Token Classification

Inference API (serverless) does not yet support peft models for this pipeline type.

Dataset used to train AmelieSchreiber/esm2_t12_35M_lora_binding_sites_v2_cp2

Collection including AmelieSchreiber/esm2_t12_35M_lora_binding_sites_v2_cp2

ESMBind (ESMB) for Protein Binding Sites

Collection

ESMBind (or ESMB) are LoRA finetuned ESM-2 models for predicting general binding sites and active sites of proteins based on sequence alone. • 14 items • Updated Oct 8, 2023 • 4