save and read mnist dataloader as one channel images code example

Example: Image loader RGB transform

While loading your images, you could use Image.open(path).convert('RGB') on all images.
If you are using ImageFolder, this functionality should be already there using the default loader.

def myloader(filename):
    return Image.open(filename).convert('RGB')


dog_test = datasets.ImageFolder('/data/dog_images/test',transform = valid_transform,loader=myloader)
test = torch.utils.data.DataLoader(dataset=dog_test,batch_size=64)