bring in ddim sampling

denoising_diffusion_pytorch/denoising_diffusion_pytorch.py

@@ -533,7 +533,7 @@ class GaussianDiffusion(nn.Module):
         return img
 
     @torch.no_grad()
-    def ddim_sample(self, shape, clip_denoised = True):
+    def ddim_sample(self, shape, clip_denoised = False):
         batch, device, total_timesteps, sampling_timesteps, eta, objective = shape[0], self.betas.device, self.num_timesteps, self.sampling_timesteps, self.ddim_sampling_eta, self.objective
 
         times = torch.linspace(0., total_timesteps, steps = sampling_timesteps + 2)[:-1]
@@ -551,9 +551,6 @@ class GaussianDiffusion(nn.Module):
 
             pred_noise, x_start, *_ = self.model_predictions(img, time_cond)
 
-            if clip_denoised:
-                x_start.clamp_(-1., 1.)
-
             c1 = eta * ((1 - alpha / alpha_next) * (1 - alpha_next) / (1 - alpha)).sqrt()
             c2 = ((1 - alpha_next) - torch.square(c1)).sqrt()
 

setup.py

@@ -3,7 +3,7 @@ from setuptools import setup, find_packages
 setup(
   name = 'denoising-diffusion-pytorch',
   packages = find_packages(),
-  version = '0.25.1',
+  version = '0.25.0',
   license='MIT',
   description = 'Denoising Diffusion Probabilistic Models - Pytorch',
   author = 'Phil Wang',