Hello! In partialconv2d.py :
if 'multi_channel' in kwargs:
self.multi_channel = kwargs['multi_channel']
kwargs.pop('multi_channel')
else:
self.multi_channel = False
Why should we set the args: multi-channel to be "True" if we want use Pconv for image inpainting? I think it has no effect when we use it for image inpainting if multi-channel is set to False. It seems like a waste, but it doesn't make any difference to the result.
Hello! In partialconv2d.py : if 'multi_channel' in kwargs: self.multi_channel = kwargs['multi_channel'] kwargs.pop('multi_channel') else: self.multi_channel = False
Why should we set the args: multi-channel to be "True" if we want use Pconv for image inpainting? I think it has no effect when we use it for image inpainting if multi-channel is set to False. It seems like a waste, but it doesn't make any difference to the result.
Please correct me if I make some wrongs.