LabelMap creation from torchio.Image

🐛Bug Sorry if I miss something here, but we used to declare LabelMap, from the torchio.Image function

To reproduce

# Your code here
import torchio as tio
sujbad = tio.Image(type=tio.LABEL, path='/mypath/mask.nii.gz')
sujok  = tio.LabelMap(path='/mypath/mask.nii.gz')

Expected behavior

I expect those 2 declaration to be equivalent (since the type can be specify in the Image constructor)

Actual behavior

type(sujbad)

torchio.data.image.Image

System info

Output of python <(curl -s https://raw.githubusercontent.com/fepegar/torchio/master/print_system.py):

Platform: Linux-4.15.0-139-generic-x86_64-with-glibc2.10
TorchIO:  0.18.39
PyTorch:  1.7.1
NumPy:    1.20.2
Python:   3.8.5 (default, Sep  4 2020, 07:30:14) 
[GCC 7.3.0]

[fepegar]

will it makes a difference somewhere ?

At the moment, I’m not sure. I would need to look into it. But please do move to ScalarImage, as in the examples and tutorials, as this is now considered the standard way and using Image for scalar images will be removed in the future.

[romainVala]

I understand, that creating a torico.Image with type=‘label’ is not a good idea, (because in other part of the code like for instance https://github.com/fepegar/torchio/blob/af86828360353dbe127d8195567ea4b9e83f238b/torchio/transforms/transform.py#L432, it will not be consider as a LabelMap

but is it also the case for intensity type ? should I also switch
tio.Image(type='intensity', path=path) to
tio.ScalarImage(path=path)

will it makes a difference somewhere ?