Closed BradZhone closed 1 year ago
Thanks @BradZhone for spotting this. I made a PR for the one line fix (https://github.com/facebookresearch/dlrm/pull/345). I'm also seeing the 0.50 AUC issue. @s4ayub or @colin2328 wrote the criteo kaggle preprocessing code in the torchrec repo so they should have an idea.
Hello @samiwilf , when I tried to run torchrec-dlrm with criteo-kaggle dataset, as following: Criteo Kaggle Display Advertising Challenge dataset usage
I got this error:
which points to: https://github.com/facebookresearch/dlrm/blob/69d22b99ec02ff868dbc1170e39686935f9d1274/torchrec_dlrm/data/dlrm_dataloader.py#L87
That may caused in the case of stage == "train" or stage == "test", so I modified this line to:
I don't know whether that's right? After that, it can run successfully however the AUC always equals to 0.5002527236938477 and can never update. Is there any solution ?