DataPoint.to_grayscale() -> DataPoint.rgb_to_grayscale (#7136)

torchvision/prototype/datapoints/_datapoint.py

@@ -230,7 +230,7 @@ class Datapoint(torch.Tensor):
     ) -> Datapoint:
         return self
 
-    def to_grayscale(self, num_output_channels: int = 1) -> Datapoint:
+    def rgb_to_grayscale(self, num_output_channels: int = 1) -> Datapoint:
         return self
 
     def adjust_brightness(self, brightness_factor: float) -> Datapoint:

torchvision/prototype/datapoints/_image.py

@@ -169,7 +169,7 @@ class Image(Datapoint):
         )
         return Image.wrap_like(self, output)
 
-    def to_grayscale(self, num_output_channels: int = 1) -> Image:
+    def rgb_to_grayscale(self, num_output_channels: int = 1) -> Image:
         output = self._F.rgb_to_grayscale_image_tensor(
             self.as_subclass(torch.Tensor), num_output_channels=num_output_channels
         )

torchvision/prototype/datapoints/_video.py

@@ -173,7 +173,7 @@ class Video(Datapoint):
         )
         return Video.wrap_like(self, output)
 
-    def to_grayscale(self, num_output_channels: int = 1) -> Video:
+    def rgb_to_grayscale(self, num_output_channels: int = 1) -> Video:
         output = self._F.rgb_to_grayscale_image_tensor(
             self.as_subclass(torch.Tensor), num_output_channels=num_output_channels
         )

torchvision/prototype/transforms/functional/_color.py

@@ -46,7 +46,7 @@ def rgb_to_grayscale(
     if torch.jit.is_scripting() or is_simple_tensor(inpt):
         return rgb_to_grayscale_image_tensor(inpt, num_output_channels=num_output_channels)
     elif isinstance(inpt, datapoints._datapoint.Datapoint):
-        return inpt.to_grayscale(num_output_channels=num_output_channels)
+        return inpt.rgb_to_grayscale(num_output_channels=num_output_channels)
     elif isinstance(inpt, PIL.Image.Image):
         return rgb_to_grayscale_image_pil(inpt, num_output_channels=num_output_channels)
     else: