Remove unnecessary lines (#8569)

* Remove unused line


---------

Co-authored-by: Sayak Paul <[email protected]>

Author: tolgacangoz

examples/community/regional_prompting_stable_diffusion.py

@@ -467,8 +467,6 @@ def make_emblist(self, prompts):
 
 
 def split_dims(xs, height, width):
-    xs = xs
-
     def repeat_div(x, y):
         while y > 0:
             x = math.ceil(x / 2)

src/diffusers/models/attention_processor.py

@@ -1112,9 +1112,7 @@ def __call__(
         key = key.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
         value = value.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
 
-        hidden_states = hidden_states = F.scaled_dot_product_attention(
-            query, key, value, dropout_p=0.0, is_causal=False
-        )
+        hidden_states = F.scaled_dot_product_attention(query, key, value, dropout_p=0.0, is_causal=False)
         hidden_states = hidden_states.transpose(1, 2).reshape(batch_size, -1, attn.heads * head_dim)
         hidden_states = hidden_states.to(query.dtype)
 

src/diffusers/models/controlnet_sd3.py

@@ -308,8 +308,6 @@ def forward(
                     "Passing `scale` via `joint_attention_kwargs` when not using the PEFT backend is ineffective."
                 )
 
-        height, width = hidden_states.shape[-2:]
-
         hidden_states = self.pos_embed(hidden_states)  # takes care of adding positional embeddings too.
         temb = self.time_text_embed(timestep, pooled_projections)
         encoder_hidden_states = self.context_embedder(encoder_hidden_states)

src/diffusers/models/unets/unet_stable_cascade.py

@@ -478,9 +478,7 @@ def custom_forward(*inputs):
                                 create_custom_forward(block), x, r_embed, use_reentrant=False
                             )
                         else:
-                            x = x = torch.utils.checkpoint.checkpoint(
-                                create_custom_forward(block), use_reentrant=False
-                            )
+                            x = torch.utils.checkpoint.checkpoint(create_custom_forward(block), use_reentrant=False)
                     if i < len(repmap):
                         x = repmap[i](x)
                 level_outputs.insert(0, x)

src/diffusers/pipelines/semantic_stable_diffusion/pipeline_semantic_stable_diffusion.py

@@ -661,7 +661,6 @@ def __call__(
                         noise_guidance_edit_tmp = torch.einsum(
                             "cb,cbijk->bijk", concept_weights_tmp, noise_guidance_edit_tmp
                         )
-                        noise_guidance_edit_tmp = noise_guidance_edit_tmp
                         noise_guidance = noise_guidance + noise_guidance_edit_tmp
 
                         self.sem_guidance[i] = noise_guidance_edit_tmp.detach().cpu()

tests/lora/test_lora_layers_sd3.py

@@ -153,7 +153,6 @@ def test_simple_inference_with_transformer_lora_save_load(self):
         pipe = self.pipeline_class(**components)
         pipe = pipe.to(torch_device)
         pipe.set_progress_bar_config(disable=None)
-        inputs = self.get_dummy_inputs(torch_device)
 
         pipe.transformer.add_adapter(transformer_config)
         self.assertTrue(check_if_lora_correctly_set(pipe.transformer), "Lora not correctly set in transformer")

tests/models/transformers/test_models_prior.py

@@ -144,9 +144,6 @@ def test_output_pretrained(self):
 class PriorTransformerIntegrationTests(unittest.TestCase):
     def get_dummy_seed_input(self, batch_size=1, embedding_dim=768, num_embeddings=77, seed=0):
         torch.manual_seed(seed)
-        batch_size = batch_size
-        embedding_dim = embedding_dim
-        num_embeddings = num_embeddings
 
         hidden_states = torch.randn((batch_size, embedding_dim)).to(torch_device)
 

tests/pipelines/stable_diffusion_2/test_stable_diffusion_attend_and_excite.py

@@ -142,7 +142,7 @@ def get_dummy_inputs(self, device, seed=0):
             generator = torch.manual_seed(seed)
         else:
             generator = torch.Generator(device=device).manual_seed(seed)
-        inputs = inputs = {
+        inputs = {
             "prompt": "a cat and a frog",
             "token_indices": [2, 5],
             "generator": generator,

tests/pipelines/stable_diffusion_adapter/test_stable_diffusion_adapter.py

@@ -538,7 +538,6 @@ def test_inference_batch_single_identical(
 
         # batchify inputs
         batched_inputs = {}
-        batch_size = batch_size
         for name, value in inputs.items():
             if name in self.batch_params:
                 # prompt is string

tests/pipelines/stable_diffusion_xl/test_stable_diffusion_xl_adapter.py

@@ -574,7 +574,6 @@ def test_inference_batch_single_identical(
 
         # batchify inputs
         batched_inputs = {}
-        batch_size = batch_size
         for name, value in inputs.items():
             if name in self.batch_params:
                 # prompt is string

tests/schedulers/test_scheduler_edm_euler.py

@@ -89,9 +89,6 @@ def test_from_save_pretrained(self):
             scheduler_config = self.get_scheduler_config()
             scheduler = scheduler_class(**scheduler_config)
 
-            sample = self.dummy_sample
-            residual = 0.1 * sample
-
             with tempfile.TemporaryDirectory() as tmpdirname:
                 scheduler.save_config(tmpdirname)
                 new_scheduler = scheduler_class.from_pretrained(tmpdirname)