Poverty Train and Val transforms

[juanelenter]

The WILDS paper claims to normalize and augment samples from the Poverty dataset in training. But this official repo and this non-official one do not apply any transformation. What is recommended?

Thanks

[sangmichaelxie]

At least as of v1.2.2, we do the augmentation:

wilds/examples/transforms.py

Line 99 in 6d96cff

def initialize_poverty_transform(is_training):

And the normalization occurs during preprocessing:

wilds/dataset_preprocessing/poverty/convert_poverty_to_npy.py

Line 35 in 4726775

normalize=True).get_batch()

However I'm not sure about how the v2 codepath works. @teetone @kohpangwei