Different Raysampler between train and val/test

[AddASecond]

I notice that in graf/transforms.py,
during train, class FlexGridRaySampler(RaySampler) is used
but during val/test， class FullRaySampler(RaySampler) is used

I wonder why use different raysampler？
what is the principle behind that opreation choice？

[huahangc]

In the paper, they had demonstrated the patch sampling is better than full sampling when in the train.