Giters

Attila94 / CIConv

Original code for "Zero-Shot Day-Night Domain Adaptation with a Physics Prior"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

I am getting 0.216 for Dark Zurich .

hassaanmahmood opened this issue · comments

commented

using python eval.py '/CIConv-main/experiments/3_segmentation/cityscapes_w.pth.tar' --invariant 'W'

Night time driving
Score Average : 0.412
But for DarkZurich
Score Average : 0.216

Although I am getting this warning, but I do not think that can be the issue
/lib/python3.7/site-packages/torchvision/transforms/functional.py:405: UserWarning: Argument interpolation should be of type InterpolationMode instead of int. Please, use InterpolationMode enum.
"Argument interpolation should be of type InterpolationMode instead of int. "
/lib/python3.7/site-packages/torchvision/transforms/functional.py:405: UserWarning: Argument interpolation should be of type InterpolationMode instead of int. Please, use InterpolationMode enum.
"Argument interpolation should be of type InterpolationMode instead of int. "
/lib/python3.7/site-packages/torchvision/transforms/functional.py:405: UserWarning: Argument interpolation should be of type InterpolationMode instead of int. Please, use InterpolationMode enum.
"Argument interpolation should be of type InterpolationMode instead of int. "
/lib/python3.7/site-packages/torchvision/transforms/functional.py:405: UserWarning: Argument interpolation should be of type InterpolationMode instead of int. Please, use InterpolationMode enum.
"Argument interpolation should be of type InterpolationMode instead of int. "

commented

First of all, thanks for your interest!

The reported performance in the paper is on the Dark Zurich test set. In order to evaluate on the test set, you should save your predictions and upload them to the evaluation server: https://competitions.codalab.org/competitions/23553. The warnings are indeed not related to this.

Hope this helps!