search

hendrycks / outlier-exposure

Deep Anomaly Detection with Outlier Exposure (ICLR 2019)
Apache License 2.0
541 stars 107 forks source link

proportion of ODD data in the training sample #8

Closed kcharanj closed 4 years ago

kcharanj commented 4 years ago

Hi, Could you please let me know the proportion of "in-distribution" and OOD data points in the training sample? Thanks

hendrycks commented 4 years ago

The code is available. Often the number of outliers in the batch are twice that of the in-distribution points, so the ratio is often 2 outliers for every in-distribution example.

On Tue, Feb 4, 2020 at 9:14 AM Charanjit Kaur Khosa < notifications@github.com> wrote:

Hi, Could you please let me know the proportion of "in-distribution" and OOD data points in the training sample? Thanks

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/hendrycks/outlier-exposure/issues/8?email_source=notifications&email_token=ACZBITXBU3DZA6JZIKTA5WLRBGO57A5CNFSM4KP2WSOKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IK6Z5DA, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACZBITSRECDGWRGG5LFAMZ3RBGO57ANCNFSM4KP2WSOA .

kcharanj commented 4 years ago

okay, thank you.