quality

Author: natolambert

src/diffusers/schedulers/scheduling_ddpm.py

@@ -274,8 +274,6 @@ def step(
             new_config["predict_epsilon"] = predict_epsilon
             self._internal_dict = FrozenDict(new_config)
 
-        t = timestep
-
         if model_output.shape[1] == sample.shape[1] * 2 and self.variance_type in ["learned", "learned_range"]:
             model_output, predicted_variance = torch.split(model_output, sample.shape[1], dim=1)
         else: