About tuning the pre-trained model for OOD detection

[hongxin001]

Hi,

In the paper, it writes: "Without assuming such knowledge, we use the maximum softmax probabilities to score anomalies and show that models which are pre-trained then tuned provide superior anomaly scores". So I want to confirm whether you fine-tune the whole model or just the classifier (the last layer). Thank you.

Best Regards, Hongxin

[hendrycks]

If I recall correctly, the whole network is modified.

On Sun, Dec 12, 2021 at 3:33 AM Hongxin Wei @.***> wrote:

Hi,

In the paper, it writes: "Without assuming such knowledge, we use the maximum softmax probabilities to score anomalies and show that models which are pre-trained then tuned provide superior anomaly scores". So I want to confirm whether you fine-tune the whole model or just the classifier (the last layer). Thank you.

Best Regards, Hongxin

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/hendrycks/pre-training/issues/3, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACZBITS3RTURNLP22FX6LN3UQSB77ANCNFSM5J4E2B3A . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.