If we parameterize xθ in terms of x0-prediction, how to achieve it?

[greens007]

Hi,
If we parameterize xθ in terms of x0-prediction, how to implement the inference of classifier-free guidance? The original paper only parameterizes xθ in terms of ε-prediction.

[TeaPearce]

Hi, I only looked at $\epsilon$ prediction here, but I imagine a similar combination could be used?