search

VITA-Group / DeblurGANv2

[ICCV 2019] "DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better" by Orest Kupyn, Tetiana Martyniuk, Junru Wu, Zhangyang Wang
Other
1.01k stars 265 forks source link

What batch size do you use when training? #93

Open thesby opened 3 years ago

thesby commented 3 years ago

I find that the batch size in config.yaml is 1, but 1 is not good for batch normalization. So I want to know the batch size to use when training.

FrankLinxzx commented 3 years ago

I find that the batch size in config.yaml is 1, but 1 is not good for batch normalization. So I want to know the batch size to use when training.

i use 8 but seems slower than 1 ,so use 1