Closed VishyTheFishy closed 1 year ago
I have been training on the CIFAR-10 dataset, and no matter how far I train, it continues to generate images which look like this:
Is this a possible issue with the image generation pipeline, or is it a problem with the model not training correctly?
I have the same problem ? Any solution ?
The training code might not be very high quality due to DDIM being mostly a sampling methods that reuses pretrained checkpoints.
I have been training on the CIFAR-10 dataset, and no matter how far I train, it continues to generate images which look like this:
Is this a possible issue with the image generation pipeline, or is it a problem with the model not training correctly?