Unify is_torchvision_v2_available with is_torchvision_available (huggingface#41227)

Signed-off-by: Yuanyuan Chen <cyyever@outlook.com>

Author: cyyever

src/transformers/models/deepseek_vl_hybrid/image_processing_deepseek_vl_hybrid.py

@@ -40,12 +40,7 @@
     validate_preprocess_arguments,
 )
 from ...processing_utils import ImagesKwargs
-from ...utils import (
-    TensorType,
-    filter_out_non_signature_kwargs,
-    is_vision_available,
-    logging,
-)
+from ...utils import TensorType, filter_out_non_signature_kwargs, is_vision_available, logging
 
 
 if is_vision_available():

src/transformers/models/deepseek_vl_hybrid/image_processing_deepseek_vl_hybrid_fast.py

@@ -21,6 +21,7 @@
 from typing import Optional, Union
 
 import torch
+from torchvision.transforms.v2 import functional as F
 
 from ...image_processing_utils_fast import (
     BaseImageProcessorFast,
@@ -42,11 +43,6 @@
 from .image_processing_deepseek_vl_hybrid import DeepseekVLHybridImageProcessorKwargs
 
 
-if is_torchvision_v2_available():
-    from torchvision.transforms.v2 import functional as F
-else:
-    from torchvision.transforms import functional as F
-
 
 @auto_docstring
 class DeepseekVLHybridImageProcessorFast(BaseImageProcessorFast):

src/transformers/models/deepseek_vl_hybrid/modeling_deepseek_vl_hybrid.py

@@ -29,11 +29,7 @@
 from ...modeling_outputs import ModelOutput
 from ...modeling_utils import PreTrainedModel
 from ...processing_utils import Unpack
-from ...utils import (
-    TransformersKwargs,
-    auto_docstring,
-    can_return_tuple,
-)
+from ...utils import TransformersKwargs, auto_docstring, can_return_tuple
 from ..auto import AutoModel
 from .configuration_deepseek_vl_hybrid import DeepseekVLHybridConfig
 

src/transformers/models/deepseek_vl_hybrid/modular_deepseek_vl_hybrid.py

@@ -16,6 +16,7 @@
 
 import torch
 import torch.nn as nn
+from torchvision.transforms.v2 import functional as F
 
 from ...cache_utils import Cache
 from ...image_processing_utils_fast import (
@@ -52,7 +53,6 @@
     auto_docstring,
     can_return_tuple,
     filter_out_non_signature_kwargs,
-    is_torchvision_v2_available,
     logging,
 )
 from ..auto import CONFIG_MAPPING, AutoConfig, AutoModel
@@ -69,12 +69,6 @@
 from ..sam.modeling_sam import SamLayerNorm, SamVisionNeck
 
 
-if is_torchvision_v2_available():
-    from torchvision.transforms.v2 import functional as F
-else:
-    from torchvision.transforms import functional as F
-
-
 logger = logging.get_logger(__name__)
 
 

src/transformers/models/dpt/image_processing_dpt_fast.py

@@ -25,6 +25,7 @@
 from typing import TYPE_CHECKING, Optional, Union
 
 import torch
+from torchvision.transforms.v2 import functional as F
 
 from ...image_processing_base import BatchFeature
 from ...image_processing_utils_fast import BaseImageProcessorFast
@@ -46,11 +47,6 @@
 if TYPE_CHECKING:
     from ...modeling_outputs import DepthEstimatorOutput
 
-if is_torchvision_v2_available():
-    from torchvision.transforms.v2 import functional as F
-else:
-    from torchvision.transforms import functional as F
-
 
 def get_resize_output_image_size(
     input_image: "torch.Tensor",

src/transformers/models/dpt/modular_dpt.py

@@ -32,7 +32,6 @@
 from ...utils import (
     TensorType,
     auto_docstring,
-    is_torchvision_v2_available,
     requires_backends,
 )
 from ..beit.image_processing_beit_fast import BeitImageProcessorFast
@@ -42,10 +41,7 @@
 if TYPE_CHECKING:
     from ...modeling_outputs import DepthEstimatorOutput
 
-if is_torchvision_v2_available():
-    from torchvision.transforms.v2 import functional as F
-else:
-    from torchvision.transforms import functional as F
+from torchvision.transforms.v2 import functional as F
 
 
 def get_resize_output_image_size(

src/transformers/models/llava_onevision/image_processing_llava_onevision_fast.py

@@ -22,6 +22,7 @@
 from typing import Optional, Union
 
 import torch
+from torchvision.transforms.v2 import functional as F
 
 from ...image_processing_utils import BatchFeature, get_patch_output_size, select_best_resolution
 from ...image_processing_utils_fast import (
@@ -44,12 +45,6 @@
 from .image_processing_llava_onevision import LlavaOnevisionImageProcessorKwargs
 
 
-if is_torchvision_v2_available():
-    from torchvision.transforms.v2 import functional as F
-else:
-    from torchvision.transforms import functional as F
-
-
 @auto_docstring
 class LlavaOnevisionImageProcessorFast(BaseImageProcessorFast):
     resample = PILImageResampling.BICUBIC

src/transformers/models/llava_onevision/modular_llava_onevision.py

@@ -18,6 +18,7 @@
 
 import torch
 from torch import nn
+from torchvision.transforms.v2 import functional as F
 
 from transformers.models.llava_next.image_processing_llava_next_fast import LlavaNextImageProcessorFast
 from transformers.models.llava_next_video.modeling_llava_next_video import (
@@ -50,17 +51,11 @@
     TensorType,
     auto_docstring,
     can_return_tuple,
-    is_torchvision_v2_available,
     logging,
 )
 from .image_processing_llava_onevision import LlavaOnevisionImageProcessorKwargs
 
 
-if is_torchvision_v2_available():
-    from torchvision.transforms.v2 import functional as F
-else:
-    from torchvision.transforms import functional as F
-
 logger = logging.get_logger(__name__)
 
 

src/transformers/models/owlv2/image_processing_owlv2_fast.py

@@ -23,6 +23,7 @@
 from typing import TYPE_CHECKING, Optional, Union
 
 import torch
+from torchvision.transforms.v2 import functional as F
 
 from ...image_processing_utils_fast import BaseImageProcessorFast, BatchFeature
 from ...image_transforms import center_to_corners_format, group_images_by_shape, reorder_images
@@ -31,12 +32,6 @@
 from .image_processing_owlv2 import _scale_boxes, box_iou
 
 
-if is_torchvision_v2_available():
-    from torchvision.transforms.v2 import functional as F
-else:
-    from torchvision.transforms import functional as F
-
-
 if TYPE_CHECKING:
     from .modeling_owlv2 import Owlv2ObjectDetectionOutput
 

src/transformers/models/owlv2/modular_owlv2.py

@@ -18,6 +18,7 @@
 from typing import Optional, Union
 
 import torch
+from torchvision.transforms.v2 import functional as F
 
 from ...image_processing_utils_fast import (
     BatchFeature,
@@ -33,17 +34,9 @@
 from ...utils import (
     TensorType,
     auto_docstring,
-    is_torchvision_v2_available,
 )
 from ..owlvit.image_processing_owlvit_fast import OwlViTImageProcessorFast
 
-
-if is_torchvision_v2_available():
-    from torchvision.transforms.v2 import functional as F
-else:
-    from torchvision.transforms import functional as F
-
-
 @auto_docstring
 class Owlv2ImageProcessorFast(OwlViTImageProcessorFast):
     resample = PILImageResampling.BILINEAR