Downsampling segmentation labels vs Upsampling predictions

Hi guys,

Thanks for having this awesome repo. I'm writing some code for semantic segmentation and noted that you are downsampling labels instead of upsampling predictions. This issue is not a code issue, but rather an open ended question:

1. Downsampling segmentation labels may lead to loss of some spatially correlated pixels, which the `NLLLoss` may not penalize, so is downsampling justified simply because it leads to less memory consumption?

2. How would we go about implementing auxiliary losses like `SELoss`[1] or patchwise pixel loss if we downsample segmentation labels?

[1]: EncNet