Output shape is BHWC instead of BCHW ?

[tcourat]

Hi,

It seems that the output tensor shape is in BHWD format : https://github.com/gastruc/AnySat/blob/e03e72e4cb52bfaffc7d0b6831da2caea7236696/src/models/networks/encoder/Any_multi.py#L261C1-L262C1 , also shown in the demo here https://github.com/gastruc/AnySat/blob/main/demo.ipynb

But in the readme it is stated that it is in BDHW format :

features = AnySat(data, patch_size=10, output='tile') #tensor of size [B, D,]
features = AnySat(data, patch_size=10, output='patch') #tensor of size [B, D, 6, 6]
features = AnySat(data, patch_size=20, output='patch') #tensor of size [B, D, 3, 3]
features = AnySat(data, patch_size=20, output='dense', output_modality='aerial') #tensor of size [B, 2*D, 30, 30]
features = AnySat(data, patch_size=20, output='dense', output_modality='s2') #tensor of size [B, 2*D, 6, 6]

So which one is the intented one ? Moreover why output BHWD format instead of BDHW which is the pytorch standard ?

Thanks