It is easy to overfit - Githubissues

jefflai108 / ASSERT

JHU's system submission to the ASVspoof 2019 Challenge: Anti-Spoofing with Squeeze-Excitation and Residual neTworks (ASSERT).

MIT License

56 stars 32 forks source link

Closed wuqiangch closed 4 years ago

wuqiangch commented 4 years ago

@jefflai108 I can get err=0.514 on PA dev set, but get err=4.312% on PA eval set. How to solve the problem? Thanks !

jefflai108 commented 4 years ago

typical techniques to prevent overfitting include: weight normalization (l1, l2 norm), dropout (most common), cross validation...