Update processing_phi3_v.py

processing_phi3_v.py

@@ -160,12 +160,12 @@ class Phi3VImageProcessor(BaseImageProcessor):
     model_input_names = ["pixel_values"]
 
     def __init__(
-            self,
-            num_crops: int = 1,
-            image_mean: Optional[Union[float, List[float]]] = None,
-            image_std: Optional[Union[float, List[float]]] = None,
-            do_convert_rgb: bool = True,
-            **kwargs,
+        self,
+        num_crops: int = 1,
+        image_mean: Optional[Union[float, List[float]]] = None,
+        image_std: Optional[Union[float, List[float]]] = None,
+        do_convert_rgb: bool = True,
+        **kwargs,
     ) -> None:
         super().__init__(**kwargs)
         self.num_crops = num_crops
@@ -174,8 +174,8 @@ class Phi3VImageProcessor(BaseImageProcessor):
         self.do_convert_rgb = do_convert_rgb
 
     def calc_num_image_tokens(
-            self,
-            images: ImageInput
+        self,
+        images: ImageInput
     ):
         """ Calculate the number of image tokens for each image.
         Args:
@@ -210,12 +210,12 @@ class Phi3VImageProcessor(BaseImageProcessor):
         return num_img_tokens
 
     def preprocess(
-            self,
-            images: ImageInput,
-            image_mean: Optional[Union[float, List[float]]] = None,
-            image_std: Optional[Union[float, List[float]]] = None,
-            do_convert_rgb: bool = None,
-            return_tensors: Optional[Union[str, TensorType]] = None,
+        self,
+        images: ImageInput,
+        image_mean: Optional[Union[float, List[float]]] = None,
+        image_std: Optional[Union[float, List[float]]] = None,
+        do_convert_rgb: bool = None,
+        return_tensors: Optional[Union[str, TensorType]] = None,
     ):
         """
         Args:
@@ -276,8 +276,7 @@ class Phi3VImageProcessor(BaseImageProcessor):
         # reshape to channel dimension -> (num_images, num_crops, 3, 336, 336)
         # (1, 3, h//336, 336, w//336, 336) -> (1, h//336, w//336, 3, 336, 336) -> (h//336*w//336, 3, 336, 336)
         hd_images_reshape = [
-            im.reshape(1, 3, h // 336, 336, w // 336, 336).permute(0, 2, 4, 1, 3, 5).reshape(-1, 3, 336,
-                                                                                             336).contiguous() for
+            im.reshape(1, 3, h // 336, 336, w // 336, 336).permute(0, 2, 4, 1, 3, 5).reshape(-1, 3, 336, 336).contiguous() for
             im, (h, w) in zip(hd_images, shapes)]
         # concat global image and local image
         hd_images_reshape = [torch.cat([_global_image] + [_im], dim=0) for _global_image, _im in
@@ -443,11 +442,9 @@ class Phi3VProcessor(ProcessorMixin):
         unique_image_ids = sorted(list(set(image_ids)))
         # image_ids must start from 1, and must be continuous int, e.g. [1, 2, 3], cannot be [1, 4, 5]
         # check the condition
-        assert unique_image_ids == list(range(1,
-                                              len(unique_image_ids) + 1)), f"image_ids must start from 1, and must be continuous int, e.g. [1, 2, 3], cannot be {unique_image_ids}"
+        assert unique_image_ids == list(range(1, len(unique_image_ids) + 1)), f"image_ids must start from 1, and must be continuous int, e.g. [1, 2, 3], cannot be {unique_image_ids}"
         # total images must be the same as the number of image tags
-        assert len(unique_image_ids) == len(
-            images), f"total images must be the same as the number of image tags, got {len(unique_image_ids)} image tags and {len(images)} images"
+        assert len(unique_image_ids) == len(images), f"total images must be the same as the number of image tags, got {len(unique_image_ids)} image tags and {len(images)} images"
 
         image_ids_pad = [[-iid] * num_img_tokens[iid - 1] for iid in image_ids]