search

DeepRec-AI / DeepRec

DeepRec is a high-performance recommendation deep learning framework based on TensorFlow. It is hosted in incubation in LF AI & Data Foundation.
Apache License 2.0
1.03k stars 349 forks source link

sparse feature importance support by DeepRec #1000

Open welsonzhang opened 3 months ago

welsonzhang commented 3 months ago

System information

Describe the feature and the current behavior/state. how deeprec analyze the sparse feature importance ? just like https://github.com/PaddlePaddle/PaddleRec/blob/master/doc/feature_importance.md https://github.com/PaddlePaddle/PaddleRec/blob/master/tools/feature_importance.py

it will shuffle sparse slots in the dataset, then eval. Comparing the baseline AUC and the AUC after slot_shuffle, the greater the AUC decrease, the more important the feature.

Will this change the current api? How?

Who will benefit with this feature? it help to analyze the sparse feature importance

Any Other info.

candyzone commented 3 months ago

Thanks for your issue. DeepRec does not support analysis of sparse feature importance. Welcome to submit design proposal to DeepRec.