about evaluation.

[twowwj]

Hi,

When I try to run the code of evaluation, I need to load the "config.pkl" as here shown:

multiple-attention/evaluation.py

Line 13 in bb069cc

with open('runs/%s/config.pkl'%name,'rb') as f:

I didn't find any evaluation config parameters in "config.py" except training parameters.
Could you give me some help?

[wangjbo]

where can we get the dataset？can you share the dataset？ My email 708873100@qq.com,thanks you very much

[sharadchandakacherla]

@yoctta please help us out, I also want to run the code in evaluation mode, however there is no celeb.json , i'm guessing this is for Celeb-DF-v1. Can you please outline steps how to evaluate the code?
Thanks

[senemaktas]

have anyone solved this? There is lack of information to run things. Thanks.