successfully did some basic math and clipped the predicted x0 intermediate for the continuous time case

denoising_diffusion_pytorch/continuous_time_gaussian_diffusion.py

@@ -59,14 +59,11 @@ class MonotonicLinear(nn.Module):
 
 # log(snr) that approximates the original linear schedule
 
-def log(t, eps = 1e-20):
-    return torch.log(t.clamp(min = eps))
-
 def beta_linear_log_snr(t):
-    return -log(expm1(1e-4 + 10 * (t ** 2)))
+    return -torch.log(expm1(1e-4 + 10 * (t ** 2)))
 
-def alpha_cosine_log_snr(t, s = 0.008):
-    return -log((torch.cos((t + s) / (1 + s) * torch.pi * 0.5) ** -2) - 1)
+def alpha_cosine_log_snr(t):
+    raise NotImplementedError
 
 class learned_noise_schedule(nn.Module):
     """ described in section H and then I.2 of the supplementary material for variational ddpm paper """
@@ -138,8 +135,6 @@ class ContinuousTimeGaussianDiffusion(nn.Module):
 
         if noise_schedule == 'linear':
             self.log_snr = beta_linear_log_snr
-        elif noise_schedule == 'cosine':
-            self.log_snr = alpha_cosine_log_snr
         elif noise_schedule == 'learned':
             log_snr_max, log_snr_min = [beta_linear_log_snr(torch.tensor([time])).item() for time in (0., 1.)]
 
@@ -192,7 +187,7 @@ class ContinuousTimeGaussianDiffusion(nn.Module):
             # in Imagen, this was changed to dynamic thresholding
             x_start.clamp_(-1., 1.)
 
-            model_mean = alpha_next * (x * (1 - c) / alpha + c * x_start)
+            model_mean = alpha_next / alpha * x * (1 - c) + alpha_next * c * x_start
         else:
             model_mean = alpha_next / alpha * (x - c * sigma * pred_noise)
 

setup.py

@@ -3,7 +3,7 @@ from setuptools import setup, find_packages
 setup(
   name = 'denoising-diffusion-pytorch',
   packages = find_packages(),
-  version = '0.17.7',
+  version = '0.17.5',
   license='MIT',
   description = 'Denoising Diffusion Probabilistic Models - Pytorch',
   author = 'Phil Wang',