huawei-noah/vega

searching valid metric is better than re-train valid metric in autoFIS?

Aperta

#107 aperta il 8 apr 2021

 (2 commenti) (0 reazioni) (1 assegnatario)Python (177 fork)auto 404
good first issuequestion

Metriche repository

Star
 (848 stelle)
Metriche merge PR
 (Metriche PR in attesa)

Descrizione

Thank you for your excellent codes. I run your autoFIS codes (autoDeepFM), but found that the valid metric after searching (auc=0.7896248002165958) is better than the retrain valid metric (auc=0.7851094769386633) in the log. i.e. [searching valid metric] > [re-train valid metric] > [official valid metric], it looks like the performance improvement is achieved by BN but autofis damages the valid metric. Something wrong ?

the valid metric after searching: image the valid metric after re-train: image

Guida contributor