search

xuebinqin / DIS

This is the repo for our new project Highly Accurate Dichotomous Image Segmentation
Apache License 2.0
2.11k stars 244 forks source link

what tar loss mean? #93

Closed babyta closed 7 months ago

babyta commented 8 months ago

Like the question, please tell my confusion

xuebinqin commented 7 months ago

because our model fuses losses from multiple decoder stages, the tar loss is the "target" loss, which indicate the loss for the last decoder stage, of which the prediction map is with the largest size.

On Sun, Nov 5, 2023 at 6:29 PM babyta @.***> wrote:

Like the question, please tell my confusion

— Reply to this email directly, view it on GitHub https://github.com/xuebinqin/DIS/issues/93, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADSGORJKM6EX5M4RSJFGL6LYDBDPDAVCNFSM6AAAAAA66ZVTFGVHI2DSMVQWIX3LMV43ASLTON2WKOZRHE3TQMJVGYYTCOI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

-- Xuebin Qin PhD Department of Computing Science University of Alberta, Edmonton, AB, Canada Homepage: https://xuebinqin.github.io/

babyta commented 7 months ago

thank you!