---------

Co-authored-by: kijai <40791699+kijai@users.noreply.github.com>
Co-authored-by: Talmaj Marinc <talmaj@comfy.org>
This commit is contained in:
Talmaj
2026-04-30 01:30:08 +02:00
committed by GitHub
parent 0e25a6936e
commit 5eeae3f1d8
11 changed files with 1345 additions and 2 deletions
+24
View File
@@ -54,6 +54,30 @@ class V_PREDICTION(EPS):
sigma = reshape_sigma(sigma, model_output.ndim)
return model_input * self.sigma_data ** 2 / (sigma ** 2 + self.sigma_data ** 2) - model_output * sigma * self.sigma_data / (sigma ** 2 + self.sigma_data ** 2) ** 0.5
class V_PREDICTION_DDPM:
"""CogVideoX v-prediction: model receives raw x_t (unscaled), predicts velocity v.
x_0 = sqrt(alpha) * x_t - sqrt(1-alpha) * v
= x_t / sqrt(sigma^2 + 1) - v * sigma / sqrt(sigma^2 + 1)
"""
def calculate_input(self, sigma, noise):
return noise
def calculate_denoised(self, sigma, model_output, model_input):
sigma = reshape_sigma(sigma, model_output.ndim)
return model_input / (sigma ** 2 + 1.0) ** 0.5 - model_output * sigma / (sigma ** 2 + 1.0) ** 0.5
def noise_scaling(self, sigma, noise, latent_image, max_denoise=False):
sigma = reshape_sigma(sigma, noise.ndim)
if max_denoise:
noise = noise * torch.sqrt(1.0 + sigma ** 2.0)
else:
noise = noise * sigma
noise += latent_image
return noise
def inverse_noise_scaling(self, sigma, latent):
return latent
class EDM(V_PREDICTION):
def calculate_denoised(self, sigma, model_output, model_input):
sigma = reshape_sigma(sigma, model_output.ndim)