Fixed Type-hints in function defintions

src/transformers/models/detr/image_processing_detr.py

@@ -1724,7 +1724,7 @@ def get_ids_area(masks, scores, dedup=False):
 
     # inspired by https://github.com/facebookresearch/detr/blob/master/models/detr.py#L258
     def post_process_object_detection(
-        self, outputs, threshold: float = 0.5, target_sizes: Union[TensorType, list[tuple]] = None
+        self, outputs, threshold: float = 0.5, target_sizes: Optional[Union[TensorType, list[tuple]]] = None
     ):
         """
         Converts the raw output of [`DetrForObjectDetection`] into final bounding boxes in (top_left_x, top_left_y,

src/transformers/models/detr/image_processing_detr_fast.py

@@ -948,7 +948,7 @@ def get_ids_area(masks, scores, dedup=False):
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.post_process_object_detection
     def post_process_object_detection(
-        self, outputs, threshold: float = 0.5, target_sizes: Union[TensorType, list[tuple]] = None
+        self, outputs, threshold: float = 0.5, target_sizes: Optional[Union[TensorType, list[tuple]]] = None
     ):
         """
         Converts the raw output of [`DetrForObjectDetection`] into final bounding boxes in (top_left_x, top_left_y,

src/transformers/models/qwen3_omni_moe/modular_qwen3_omni_moe.py

@@ -2673,9 +2673,9 @@ def replace_multimodal_special_tokens(
     def __call__(
         self,
         text: TextInput = None,
-        images: ImageInput = None,
-        videos: VideoInput = None,
-        audio: AudioInput = None,
+        images: Optional[ImageInput] = None,
+        videos: Optional[VideoInput] = None,
+        audio: Optional[AudioInput] = None,
         **kwargs,
     ):
         """

src/transformers/models/qwen3_omni_moe/processing_qwen3_omni_moe.py

@@ -20,7 +20,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import re
-from typing import Union
+from typing import Optional, Union
 
 import numpy as np
 
@@ -123,9 +123,9 @@ def __init__(
     def __call__(
         self,
         text: TextInput = None,
-        images: ImageInput = None,
-        videos: VideoInput = None,
-        audio: AudioInput = None,
+        images: Optional[ImageInput] = None,
+        videos: Optional[VideoInput] = None,
+        audio: Optional[AudioInput] = None,
         **kwargs,
     ) -> BatchFeature:
         """

src/transformers/models/rag/modeling_rag.py

@@ -241,7 +241,7 @@ def from_pretrained_question_encoder_generator(
         cls,
         question_encoder_pretrained_model_name_or_path: Optional[str] = None,
         generator_pretrained_model_name_or_path: Optional[str] = None,
-        retriever: RagRetriever = None,
+        retriever: Optional[RagRetriever] = None,
         **kwargs,
     ) -> PreTrainedModel:
         r"""

src/transformers/models/rt_detr/image_processing_rt_detr.py

@@ -978,7 +978,7 @@ def post_process_object_detection(
         self,
         outputs,
         threshold: float = 0.5,
-        target_sizes: Union[TensorType, list[tuple]] = None,
+        target_sizes: Optional[Union[TensorType, list[tuple]]] = None,
         use_focal_loss: bool = True,
     ):
         """

src/transformers/models/sam2_video/modular_sam2_video.py

@@ -635,9 +635,9 @@ def init_video_session(
         self,
         video: Optional[VideoInput] = None,
         inference_device: Union[str, "torch.device"] = "cpu",
-        inference_state_device: Union[str, "torch.device"] = None,
-        processing_device: Union[str, "torch.device"] = None,
-        video_storage_device: Union[str, "torch.device"] = None,
+        inference_state_device: Optional[Union[str, "torch.device"]] = None,
+        processing_device: Optional[Union[str, "torch.device"]] = None,
+        video_storage_device: Optional[Union[str, "torch.device"]] = None,
         max_vision_features_cache_size: int = 1,
         dtype: torch.dtype = torch.float32,
     ):

src/transformers/models/sam2_video/processing_sam2_video.py

@@ -530,9 +530,9 @@ def init_video_session(
         self,
         video: Optional[VideoInput] = None,
         inference_device: Union[str, "torch.device"] = "cpu",
-        inference_state_device: Union[str, "torch.device"] = None,
-        processing_device: Union[str, "torch.device"] = None,
-        video_storage_device: Union[str, "torch.device"] = None,
+        inference_state_device: Optional[Union[str, "torch.device"]] = None,
+        processing_device: Optional[Union[str, "torch.device"]] = None,
+        video_storage_device: Optional[Union[str, "torch.device"]] = None,
         max_vision_features_cache_size: int = 1,
         dtype: torch.dtype = torch.float32,
     ):

src/transformers/models/seamless_m4t/tokenization_seamless_m4t.py

@@ -211,9 +211,9 @@ def vocab_size(self):
 
     def __call__(
         self,
-        text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]] = None,
+        text: Optional[Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]] = None,
         text_pair: Optional[Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]] = None,
-        text_target: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]] = None,
+        text_target: Optional[Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]] = None,
         text_pair_target: Optional[
             Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]
         ] = None,

src/transformers/models/seamless_m4t/tokenization_seamless_m4t_fast.py

@@ -371,9 +371,9 @@ def _from_pretrained(
 
     def __call__(
         self,
-        text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]] = None,
+        text: Optional[Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]] = None,
         text_pair: Optional[Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]] = None,
-        text_target: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]] = None,
+        text_target: Optional[Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]] = None,
         text_pair_target: Optional[
             Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]
         ] = None,

src/transformers/models/trocr/processing_trocr.py

@@ -53,7 +53,7 @@ def __init__(self, image_processor=None, tokenizer=None, **kwargs):
     def __call__(
         self,
         images: Optional[ImageInput] = None,
-        text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]] = None,
+        text: Optional[Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]] = None,
         **kwargs: Unpack[TrOCRProcessorKwargs],
     ) -> BatchFeature:
         """

src/transformers/models/udop/tokenization_udop.py

@@ -508,11 +508,11 @@ def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] =
     @add_end_docstrings(UDOP_ENCODE_KWARGS_DOCSTRING)
     def __call__(
         self,
-        text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]] = None,
+        text: Optional[Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]] = None,
         text_pair: Optional[Union[PreTokenizedInput, list[PreTokenizedInput]]] = None,
         boxes: Optional[Union[list[list[int]], list[list[list[int]]]]] = None,
         word_labels: Optional[Union[list[int], list[list[int]]]] = None,
-        text_target: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]] = None,
+        text_target: Optional[Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]] = None,
         text_pair_target: Optional[
             Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]
         ] = None,
@@ -703,7 +703,7 @@ def batch_encode_plus_boxes(
         word_labels: Optional[list[list[int]]] = None,
         add_special_tokens: bool = True,
         padding: Union[bool, str, PaddingStrategy] = False,
-        truncation: Union[bool, str, TruncationStrategy] = None,
+        truncation: Optional[Union[bool, str, TruncationStrategy]] = None,
         max_length: Optional[int] = None,
         stride: int = 0,
         is_split_into_words: bool = False,
@@ -771,7 +771,7 @@ def encode_boxes(
         word_labels: Optional[list[list[int]]] = None,
         add_special_tokens: bool = True,
         padding: Union[bool, str, PaddingStrategy] = False,
-        truncation: Union[bool, str, TruncationStrategy] = None,
+        truncation: Optional[Union[bool, str, TruncationStrategy]] = None,
         max_length: Optional[int] = None,
         stride: int = 0,
         return_tensors: Optional[Union[str, TensorType]] = None,
@@ -814,7 +814,7 @@ def encode_plus_boxes(
         word_labels: Optional[list[list[int]]] = None,
         add_special_tokens: bool = True,
         padding: Union[bool, str, PaddingStrategy] = False,
-        truncation: Union[bool, str, TruncationStrategy] = None,
+        truncation: Optional[Union[bool, str, TruncationStrategy]] = None,
         max_length: Optional[int] = None,
         stride: int = 0,
         is_split_into_words: bool = False,

src/transformers/models/video_llava/processing_video_llava.py

@@ -93,8 +93,8 @@ def __call__(
         images: Optional[ImageInput] = None,
         videos: Optional[ImageInput] = None,
         padding: Union[bool, str, PaddingStrategy] = False,
-        truncation: Union[bool, str, TruncationStrategy] = None,
-        max_length=None,
+        truncation: Optional[Union[bool, str, TruncationStrategy]] = None,
+        max_length: Optional[int] = None,
         return_tensors: Optional[Union[str, TensorType]] = TensorType.PYTORCH,
     ) -> BatchFeature:
         """

src/transformers/models/vitpose/image_processing_vitpose.py

@@ -595,7 +595,7 @@ def post_process_pose_estimation(
         boxes: Union[list[list[list[float]]], np.ndarray],
         kernel_size: int = 11,
         threshold: Optional[float] = None,
-        target_sizes: Union[TensorType, list[tuple]] = None,
+        target_sizes: Optional[Union[TensorType, list[tuple]]] = None,
     ):
         """
         Transform the heatmaps into keypoint predictions and transform them back to the image.

src/transformers/models/wav2vec2/modeling_wav2vec2.py

@@ -1474,7 +1474,7 @@ def compute_contrastive_logits(
         target_features: torch.FloatTensor,
         negative_features: torch.FloatTensor,
         predicted_features: torch.FloatTensor,
-        temperature: int = 0.1,
+        temperature: float = 0.1,
     ):
         """
         Compute logits for contrastive loss based using cosine similarity as the distance measure between

src/transformers/models/wav2vec2_conformer/modeling_wav2vec2_conformer.py

@@ -1228,7 +1228,7 @@ def compute_contrastive_logits(
         target_features: torch.FloatTensor,
         negative_features: torch.FloatTensor,
         predicted_features: torch.FloatTensor,
-        temperature: int = 0.1,
+        temperature: float = 0.1,
     ):
         """
         Compute logits for contrastive loss based using cosine similarity as the distance measure between

src/transformers/models/xcodec/configuration_xcodec.py

@@ -98,8 +98,8 @@ def __init__(
         codebook_size: int = 1024,
         codebook_dim: Optional[int] = None,
         initializer_range: float = 0.02,
-        acoustic_model_config: Union[dict, DacConfig] = None,
-        semantic_model_config: Union[dict, HubertConfig] = None,
+        acoustic_model_config: Optional[Union[dict, DacConfig]] = None,
+        semantic_model_config: Optional[Union[dict, HubertConfig]] = None,
         **kwargs,
     ):
         if acoustic_model_config is None:

src/transformers/models/yolos/image_processing_yolos.py

@@ -1471,7 +1471,7 @@ def post_process(self, outputs, target_sizes):
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.post_process_object_detection with Detr->Yolos
     def post_process_object_detection(
-        self, outputs, threshold: float = 0.5, target_sizes: Union[TensorType, list[tuple]] = None
+        self, outputs, threshold: float = 0.5, target_sizes: Optional[Union[TensorType, list[tuple]]] = None
     ):
         """
         Converts the raw output of [`YolosForObjectDetection`] into final bounding boxes in (top_left_x, top_left_y,

src/transformers/utils/quantization_config.py

@@ -1128,11 +1128,11 @@ def __init__(
         in_features: int = -1,
         indices_as_float: bool = False,
         is_indice_packed: bool = True,
-        num_centroids: tuple = [-1, -1],
-        num_res_centroids: tuple = [-1, -1],
+        num_centroids: list = [-1, -1],
+        num_res_centroids: list = [-1, -1],
         out_features: int = -1,
         outlier_size: int = 0,
-        vector_lens: tuple = [-1, -1],
+        vector_lens: list = [-1, -1],
         **kwargs,
     ):
         self.enable_norm = enable_norm

tests/models/tvp/test_image_processing_tvp.py

@@ -47,7 +47,7 @@ def __init__(
         do_pad: bool = True,
         pad_size: dict[str, int] = {"height": 80, "width": 80},
         fill: Optional[int] = None,
-        pad_mode: PaddingMode = None,
+        pad_mode: Optional[PaddingMode] = None,
         do_normalize: bool = True,
         image_mean: Optional[Union[float, list[float]]] = [0.48145466, 0.4578275, 0.40821073],
         image_std: Optional[Union[float, list[float]]] = [0.26862954, 0.26130258, 0.27577711],