val_loss is not very good in my keras implementation.

zhiguowang / BiMPM

BiMPM: Bilateral Multi-Perspective Matching for Natural Language Sentences

Apache License 2.0

438 stars 150 forks source link

val_loss is not very good in my keras implementation. #6

Open ijinmao opened 7 years ago

ijinmao commented 7 years ago

I implement a simplified keras version here. However the val_loss is not very good compared to a simple lstm network without matching layer. Could you help me to check it out if my implementation has any bugs? Thanks.