0.0.51

suppress a warning

0.0.50

no CLIP altogether for training DiffusionPrior

0.0.49

have researcher explicitly state upfront whether to condition with te…

0.0.48

allow for training the Prior network with precomputed CLIP embeddings…

0.0.47

refactor so that the causal transformer in the diffusion prior networ…

0.0.46

make sure non-latent diffusion still works

0.0.45

start a file for all attention-related modules, use attention-based u…

0.0.44

bring in attention-based upsampling to strengthen vqgan-vae, seems to…

0.0.43

better naming

0.0.42

just take care of the logic for setting all latent diffusion to predi…