Help needed in kaggle competition

[abhigupta4981 (Abhi Gupta)]

I was trying this kaggle competition. A kernel suggested that we should convert the given images into one channel and save them to a ‘.npz’ file. How to work with this type of file using fastai_v1? Any help appreciated

[mnpinto (Miguel Pinto)]

There is a kernel using fastai, I think it’s not v1 but you probably can take some ideas from there. And you should use the discussion section on kaggle for competition related questions.

[abhigupta4981 (Abhi Gupta)]

ok thanks, will heed to your advice

[KarlH (Karl)]

You could also save the images as 3 channel RGB images and plug that right into fastai

[ste (Stefano Giomo)]

Same here:

ImageData.get_ds in fastai v1?

Hi @edwincv0, I was trying to address the same competition, according to the suggestion I gave to you, but I got stuck integrating the loader in a DataBunch… This is the multi-channel image loader that loads n images and build a single multi channel image: import PIL def openMultiChannelImage(fpArr): ''' Open multiple images and return a single multi channel image ''' mat = None nChannels = len(fpArr) for i,fp in enumerate(fpArr): #print('Loading: ', fp) …

[Bibek (Bibek Chaudhary)]

I am also working on the same competition to apply the learnings of Lesson-2 but got stuck here. Any help?