Few more changes to solve some other pre-commit hooks failures

Signed-off-by: Christian Pinto <christian.pinto@ibm.com>

Author: christian-pinto

tests/models/multimodal/pooling/test_prithvi_mae.py

@@ -6,31 +6,35 @@
 
 from ....conftest import VllmRunner
 
+
 def generate_test_mm_data():
     mm_data = {
         "pixel_values": torch.full((6, 512, 512), 1.0, dtype=torch.float16),
         "location_coords": torch.full((1, 2), 1.0, dtype=torch.float16),
     }
     return mm_data
-     
+
+
 def _run_test(
     vllm_runner: type[VllmRunner],
     model: str,
-) -> None:   
+) -> None:
 
     mm_data = generate_test_mm_data()
     prompt = {
         # This model deals with no text input
         "prompt_token_ids": [1],
         "multi_modal_data": mm_data
     }
-    with vllm_runner(model, task="embed",
-                    dtype=torch.float16,
-                    enforce_eager=True,
-                    skip_tokenizer_init=True) as vllm_model:
-        output = vllm_model.encode(prompt)
-            
-MODELS=["christian-pinto/Prithvi-EO-2.0-300M-TL-VLLM"]
+    with vllm_runner(model,
+                     task="embed",
+                     dtype=torch.float16,
+                     enforce_eager=True,
+                     skip_tokenizer_init=True) as vllm_model:
+        vllm_model.encode(prompt)
+
+MODELS = ["christian-pinto/Prithvi-EO-2.0-300M-TL-VLLM"]
+
 @pytest.mark.parametrize("model", MODELS)
 def test_models_image(
     hf_runner,
@@ -41,4 +45,4 @@ def test_models_image(
     _run_test(
         vllm_runner,
         model,
-    )
+    )

vllm/v1/core/kv_cache_manager.py

@@ -1,6 +1,7 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
+from abc import ABC, abstractmethod
 from collections import defaultdict
 from dataclasses import dataclass
 from typing import Optional
@@ -65,7 +66,6 @@ def new_empty(self) -> "KVCacheBlocks":
 
 
 class KVCacheManager:
-
     def __init__(
         self,
         kv_cache_config: KVCacheConfig,

vllm/v1/core/sched/scheduler.py

@@ -488,8 +488,8 @@ def schedule(self) -> SchedulerOutput:
 
                 if self.lora_config and request.lora_request:
                     scheduled_loras.add(request.lora_request.lora_int_id)
-                req_to_new_block_ids[request.request_id] = (
-                    self.kv_cache_manager.get_block_ids(request.request_id))
+                req_to_new_block_ids[request.request_id] = \
+                    self.kv_cache_manager.get_block_ids(request.request_id)
                 num_scheduled_tokens[request.request_id] = num_new_tokens
                 token_budget -= num_new_tokens
                 request.status = RequestStatus.RUNNING