You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Not the author but was looking through issues here. You are calling on the "classifier_sample" script. Which samples from an existing classifier model.
To train, take a look at "classifier_train"
As for what parameters are necessary, there is a lot of default options set, so it depends on what you want. Seems like you are aiming for a 128x128 image, and that's a parameter you would have to set.
May I ask if the MODEL FLAGS parameter here can be used to train the diffusion model? Is only one parameter needed for training the diffusion model?
The text was updated successfully, but these errors were encountered: