Reading an alpha channel into a Data Bunch

[ste (Stefano Giomo)]

Very Interesting!

@csaroff

ImageList.from_folder(path_img, convert_mode=‘RGBA’)

There are two steps to use a RGBA image:

1. Properly load it
2. Adapt the first layer of your model to accept 4 channels instead of the canonical 3.

See

How to work with 4+ channel images in fastai_v1

These are two examples of how to deal with multiple channels images (both based on human atlas competition): https://forums.fast.ai/t/human-protein-atlas-competition-starter-code/29807?u=ste https://forums.fast.ai/t/classifier-with-multiple-images-as-input-and-multiple-labels-as-output/29704/5?u=ste The current version of data block api is very simple to customize: my code that use 4+ channel images is: ... class MultiChannelImageItemList(ImageItemList): def open(self, fn): …