Apply suggestions from code review

Author: yiyixuxu

src/diffusers/guiders/adaptive_projected_guidance.py

@@ -81,7 +81,7 @@ def prepare_inputs(self, data: Dict[str, Tuple[torch.Tensor, torch.Tensor]]) ->
         if self._step == 0:
             if self.adaptive_projected_guidance_momentum is not None:
                 self.momentum_buffer = MomentumBuffer(self.adaptive_projected_guidance_momentum)
-        tuple_indices = [0] if self.num_conditions == 1 or not self._is_apg_enabled() else [0, 1]
+        tuple_indices = [0] if self.num_conditions == 1 else [0, 1]
         data_batches = []
         for tuple_idx, input_prediction in zip(tuple_indices, self._input_predictions):
             data_batch = self._prepare_batch(data, tuple_idx, input_prediction)

src/diffusers/guiders/adaptive_projected_guidance_mix.py

@@ -93,7 +93,7 @@ def prepare_inputs(self, data: Dict[str, Tuple[torch.Tensor, torch.Tensor]]) ->
             if self.adaptive_projected_guidance_momentum is not None:
                 self.momentum_buffer = MomentumBuffer(self.adaptive_projected_guidance_momentum)
         tuple_indices = (
-            [0] if self.num_conditions == 1 or not self._is_apg_enabled() and not self._is_cfg_enabled() else [0, 1]
+            [0] if self.num_conditions == 1 else [0, 1]
         )
         data_batches = []
         for tuple_idx, input_prediction in zip(tuple_indices, self._input_predictions):

src/diffusers/guiders/auto_guidance.py

@@ -134,7 +134,7 @@ def cleanup_models(self, denoiser: torch.nn.Module) -> None:
                 registry.remove_hook(name, recurse=True)
 
     def prepare_inputs(self, data: Dict[str, Tuple[torch.Tensor, torch.Tensor]]) -> List["BlockState"]:
-        tuple_indices = [0] if self.num_conditions == 1 or not self._is_ag_enabled() else [0, 1]
+        tuple_indices = [0] if self.num_conditions == 1 else [0, 1]
         data_batches = []
         for tuple_idx, input_prediction in zip(tuple_indices, self._input_predictions):
             data_batch = self._prepare_batch(data, tuple_idx, input_prediction)

src/diffusers/guiders/classifier_free_guidance.py

@@ -92,7 +92,7 @@ def __init__(
         self.use_original_formulation = use_original_formulation
 
     def prepare_inputs(self, data: Dict[str, Tuple[torch.Tensor, torch.Tensor]]) -> List["BlockState"]:
-        tuple_indices = [0] if self.num_conditions == 1 or not self._is_cfg_enabled() else [0, 1]
+        tuple_indices = [0] if self.num_conditions == 1 else [0, 1]
         data_batches = []
         for tuple_idx, input_prediction in zip(tuple_indices, self._input_predictions):
             data_batch = self._prepare_batch(data, tuple_idx, input_prediction)

src/diffusers/guiders/classifier_free_zero_star_guidance.py

@@ -78,7 +78,7 @@ def __init__(
         self.use_original_formulation = use_original_formulation
 
     def prepare_inputs(self, data: Dict[str, Tuple[torch.Tensor, torch.Tensor]]) -> List["BlockState"]:
-        tuple_indices = [0] if self.num_conditions == 1 or not self._is_cfg_enabled() else [0, 1]
+        tuple_indices = [0] if self.num_conditions == 1 else [0, 1]
         data_batches = []
         for tuple_idx, input_prediction in zip(tuple_indices, self._input_predictions):
             data_batch = self._prepare_batch(data, tuple_idx, input_prediction)

src/diffusers/guiders/frequency_decoupled_guidance.py

@@ -219,7 +219,7 @@ def __init__(
             )
 
     def prepare_inputs(self, data: Dict[str, Tuple[torch.Tensor, torch.Tensor]]) -> List["BlockState"]:
-        tuple_indices = [0] if self.num_conditions == 1 or not self._is_fdg_enabled() else [0, 1]
+        tuple_indices = [0] if self.num_conditions == 1 else [0, 1]
         data_batches = []
         for tuple_idx, input_prediction in zip(tuple_indices, self._input_predictions):
             data_batch = self._prepare_batch(data, tuple_idx, input_prediction)

src/diffusers/guiders/perturbed_attention_guidance.py

@@ -170,7 +170,7 @@ def cleanup_models(self, denoiser: torch.nn.Module) -> None:
 
     # Copied from diffusers.guiders.skip_layer_guidance.SkipLayerGuidance.prepare_inputs
     def prepare_inputs(self, data: Dict[str, Tuple[torch.Tensor, torch.Tensor]]) -> List["BlockState"]:
-        if self.num_conditions == 1 or not self._is_cfg_enabled() and not self._is_slg_enabled():
+        if self.num_conditions == 1:
             tuple_indices = [0]
             input_predictions = ["pred_cond"]
         elif self.num_conditions == 2:

src/diffusers/guiders/skip_layer_guidance.py

@@ -166,7 +166,7 @@ def cleanup_models(self, denoiser: torch.nn.Module) -> None:
                 registry.remove_hook(hook_name, recurse=True)
 
     def prepare_inputs(self, data: Dict[str, Tuple[torch.Tensor, torch.Tensor]]) -> List["BlockState"]:
-        if self.num_conditions == 1 or not self._is_cfg_enabled() and not self._is_slg_enabled():
+        if self.num_conditions == 1:
             tuple_indices = [0]
             input_predictions = ["pred_cond"]
         elif self.num_conditions == 2:

src/diffusers/guiders/smoothed_energy_guidance.py

@@ -155,7 +155,7 @@ def cleanup_models(self, denoiser: torch.nn.Module):
                 registry.remove_hook(hook_name, recurse=True)
 
     def prepare_inputs(self, data: Dict[str, Tuple[torch.Tensor, torch.Tensor]]) -> List["BlockState"]:
-        if self.num_conditions == 1 or not self._is_cfg_enabled() and not self._is_seg_enabled():
+        if self.num_conditions == 1:
             tuple_indices = [0]
             input_predictions = ["pred_cond"]
         elif self.num_conditions == 2:

src/diffusers/guiders/tangential_classifier_free_guidance.py

@@ -67,7 +67,7 @@ def __init__(
         self.use_original_formulation = use_original_formulation
 
     def prepare_inputs(self, data: Dict[str, Tuple[torch.Tensor, torch.Tensor]]) -> List["BlockState"]:
-        tuple_indices = [0] if self.num_conditions == 1 or not self._is_tcfg_enabled() else [0, 1]
+        tuple_indices = [0] if self.num_conditions == 1 else [0, 1]
         data_batches = []
         for tuple_idx, input_prediction in zip(tuple_indices, self._input_predictions):
             data_batch = self._prepare_batch(data, tuple_idx, input_prediction)