appease linter

Signed-off-by: Max de Bayser <[email protected]>

vllm/config.py

@@ -5,19 +5,19 @@
                     Mapping, Optional, Set, Tuple, Type, Union)
 
 import torch
+from transformers import PretrainedConfig
 
 import vllm.envs as envs
-from transformers import PretrainedConfig
 from vllm.logger import init_logger
-from vllm.model_executor.layers.pooler import PoolingConfig  # noqa: F401
+from vllm.model_executor.layers.pooler import PoolingConfig
 from vllm.model_executor.layers.quantization import QUANTIZATION_METHODS
 from vllm.model_executor.models import ModelRegistry
 from vllm.platforms import current_platform
 from vllm.tracing import is_otel_available, otel_import_error_traceback
-from vllm.transformers_utils.config import get_pooling_config  # noqa: F401
 from vllm.transformers_utils.config import (ConfigFormat, get_config,
                                             get_hf_image_processor_config,
-                                            get_hf_text_config)
+                                            get_hf_text_config,
+                                            get_pooling_config)
 from vllm.utils import (GiB_bytes, cuda_device_count_stateless, get_cpu_memory,
                         is_hip, is_openvino, is_xpu, print_warning_once)
 

vllm/model_executor/model_loader/loader.py

@@ -18,17 +18,17 @@
 import torch
 from huggingface_hub import HfApi, hf_hub_download
 from torch import nn
+from transformers import AutoModelForCausalLM, PretrainedConfig
 from transformers.utils import SAFE_WEIGHTS_INDEX_NAME
 
-from transformers import AutoModelForCausalLM, PretrainedConfig
 from vllm.config import (CacheConfig, DeviceConfig, LoadConfig, LoadFormat,
                          LoRAConfig, ModelConfig, MultiModalConfig,
                          ParallelConfig, SchedulerConfig)
 from vllm.distributed import (get_tensor_model_parallel_rank,
                               get_tensor_model_parallel_world_size)
 from vllm.envs import VLLM_USE_MODELSCOPE
 from vllm.logger import init_logger
-from vllm.model_executor.layers.pooler import PoolingConfig  # noqa: F401
+from vllm.model_executor.layers.pooler import PoolingConfig
 from vllm.model_executor.layers.quantization.base_config import (
     QuantizationConfig)
 from vllm.model_executor.model_loader.tensorizer import (

vllm/model_executor/models/bert.py

@@ -2,8 +2,8 @@
 
 import torch
 from torch import nn
-
 from transformers import BertConfig
+
 from vllm.attention import Attention, AttentionMetadata, AttentionType
 from vllm.attention.backends.xformers import XFormersImpl
 from vllm.config import CacheConfig
@@ -12,8 +12,7 @@
 from vllm.model_executor.layers.linear import (ColumnParallelLinear,
                                                QKVParallelLinear,
                                                RowParallelLinear)
-from vllm.model_executor.layers.pooler import (Pooler,  # noqa: F401
-                                               PoolingConfig)
+from vllm.model_executor.layers.pooler import Pooler, PoolingConfig
 from vllm.model_executor.layers.quantization.base_config import (
     QuantizationConfig)
 from vllm.model_executor.layers.vocab_parallel_embedding import (

vllm/transformers_utils/config.py

@@ -6,13 +6,13 @@
 import huggingface_hub
 from huggingface_hub import (file_exists, hf_hub_download,
                              try_to_load_from_cache)
+from transformers import GenerationConfig, PretrainedConfig
 from transformers.models.auto.image_processing_auto import (
     get_image_processor_config)
 from transformers.models.auto.modeling_auto import (
     MODEL_FOR_CAUSAL_LM_MAPPING_NAMES)
 from transformers.utils import CONFIG_NAME as HF_CONFIG_NAME
 
-from transformers import GenerationConfig, PretrainedConfig
 from vllm.envs import VLLM_USE_MODELSCOPE
 from vllm.logger import init_logger
 # yapf conflicts with isort for this block