Val loss doesn't decrease, when train loss changes

FunctionLab / sei-framework

code to run sei and obtain sei and sequence class predictions

Other

90 stars 6 forks source link

Val loss doesn't decrease, when train loss changes #28

Open Jingyao711 opened 2 months ago

Jingyao711 commented 2 months ago

This is my config file 1720457543497 I got the model to train with our data, but the result is a bit wired. The train loss is decreasing but the val loss doesn't change much and the roc_auc value is constant. I am wondering maybe I didn't set the val correctly. Can you tell me where did I make mistake?