search

mattpoggi / mono-uncertainty

CVPR 2020 - On the uncertainty of self-supervised monocular depth estimation
233 stars 24 forks source link

What is the initial learning rate for the snap model? #10

Closed kishore-greddy closed 3 years ago

kishore-greddy commented 3 years ago

Hey @mattpoggi ,

I tried to look for initial learning rate for the "snap" training mode, but I could not find it. A couple of questions related to snap, 1) The original Snap authors use initial learning rates of 0.2 and 0.1, Did you also use 0.2 as the initial learning rate? 2) You use "C" as 20 meaning one Snap every epoch and choose last N Snaps or N random Snaps? (N being 8 according to your paper), Also T = total number of training iterations , which means for eigen zhou split it would be (39810//batch_size)* num_epochs?

Thanks in advance 👍

mattpoggi commented 3 years ago

Hi @kishore-greddy ,

  1. We used the original learning rate from monodepth2
  2. We keep the last N snaps. Your definition of T is correct
kishore-greddy commented 3 years ago

Thank you