0.0.75

fix all issues with text encodings conditioning in the decoder, using…

0.0.74

fix decoder needing separate conditional dropping probabilities for i…

0.0.73

use @clip-anytorch , thanks to @rom1504

0.0.72

fix a bug with classifier free guidance, thanks to @xiankgx again!

0.0.70

just take care of the logic for AdamW and transformers

0.0.67

now completely OpenAI CLIP compatible for training

0.0.65

fix another bug thanks to @xiankgx

0.0.64

force first unet in the cascade to be conditioned on image embeds

0.0.63

vqgan-vae codebook dims should be 256 or smaller

0.0.62

🐛