-
Hello everyone,
For the "DiffusionPriorNetwork" "forward_with_cond_scale" method:
`null_logits = self.forward(*args, text_cond_drop_prob = 1., image_cond_drop_prob = 1, **kwargs)`
Could someone …
-
in the paper , the latex is
, [2](https://github.com/Amblyopius/Stable-Di…
-
Hi, could I ask a simple question about the classifier-free guidance on class-conditioned models?
According to the original paper, we need to jointly train the conditioned and unconditioned models wi…
-
Thank you for the fantastic work and for making the code open-source. I have a small question regarding the `forward_with_cfg` function.
```
def forward_with_cfg(self, x, t, y, cfg_scale):
…
-
## 🚀 Feature
Add an option for models to use Classifier-Free Guidance (CFG) during inference. CFG uses a negative prompt to push inference to follow the system prompt more closely.
Support for thi…
-
A technique for making prompt generation stay on target used to only be applicable for images - but now can be used for text. Considering the issue for Airo 2.0, this might help. Apparently, it has…
-
Why the unconditional predicted noise didn't use referencenet feature? Which may cause a gap between training and inference. When in training, we only dropout the hiddenstates, not the referencenet fe…
-
Hi, thank you for the insightful work!
I have some concerns regarding the classifier-free guidance (CFG) in auto-regressive models.
CFG in this work is implemented as follows:
https://github.…
-
I've seen there's recently been a [PR](https://github.com/huggingface/transformers/pull/24654) that incorporates classifier-free guidance to HF model.generate function and it's super useful for genera…