！help
Closed this issue · 3 comments
Something going wrong!Look the picture below.Have a nice day!
It looks like you're trying to read in grayscale (i.e., single-channel) images, while the data class, transforms and model were written for RGB (i.e., 3-channel) images. Easiest solution, you could just repeat the image three times to create a 3-channel input.
But you'll have to play around with it a bit yourself.
You are kind and patient.But It is still difficult to me.I can’t seem to find the crux of the problem.That's my picture input and has already repeated three times.Can you give me more detailed advice?
I didn't mean you should copy it in your file system. I meant that you should transform grayscale images to RGB images on the fly. You can do that, e.g., in this line: https://github.com/mkirchler/transferGWAS/blob/master/feature_condensation/feature_condensation.py#L383 by replacing it with something like img = Image.open(p).convert('RGB') (not tested).
Also, just a heads up - you will need many more data points than ten for any GWAS to work reasonably well (or the LMM software even to run).