Table 4 Comparison of the performance of loss functions in three training data sets.

From: Accurate prediction of protein–ATP binding sites based on a protein pretrained large language model and a fractional-order convolutional neural network

Loss functions-227

Acc

Sen

Spe

Pre

MCC

AUC

Cross-entropy

0.990

0.266

0.996

0.617

0.433

0.943

Total loss(L + FL)

0.990

0.391

0.997

0.642

0.497

0.965

Loss functions-388

Acc

Sen

Spe

Pre

MCC

AUC

Cross-entropy

0.982

0.164

0.998

0.745

0.294

0.935

Total loss(L + FL)

0.984

0.194

0.999

0.857

0.403

0.941

Loss functions-1930

Acc

Sen

Spe

Pre

MCC

AUC

Cross-entropy

0.984

0.378

0.997

0.764

0.541

0.947

Total loss(L + FL)

0.986

0.385

0.998

0.775

0.540

0.953

Back to article page