[fbsync] Updated _type_conversion.py docs (#7324)

NicolasHug · facebook-github-bot · commit 0c2280ec1cd2 · 2023-03-30T06:21:15.000-07:00
Summary: Co-authored-by: Nicolas Hug &lt;contact@nicolas-hug.com&gt;

Reviewed By: vmoens

Differential Revision: D44416589

fbshipit-source-id: 801b13b054adb7466344fc46f1c165627a60629e
diff --git a/docs/source/transforms.rst b/docs/source/transforms.rst
@@ -209,6 +209,7 @@ Conversion
     v2.ToTensor
     PILToTensor
     v2.PILToTensor
+    v2.ToImageTensor
     ConvertImageDtype
     v2.ConvertImageDtype
     v2.ConvertDtype
diff --git a/torchvision/transforms/v2/_type_conversion.py b/torchvision/transforms/v2/_type_conversion.py
@@ -27,6 +27,13 @@ def _transform(self, inpt: PIL.Image.Image, params: Dict[str, Any]) -> torch.Ten
 
 
 class ToImageTensor(Transform):
+    """[BETA] Convert a tensor or an ndarray or PIL Image to :class:`~torchvision.datapoints.Image`.
+
+    .. betastatus:: ToImageTensor transform
+
+    This transform does not support torchscript.
+    """
+
     _transformed_types = (is_simple_tensor, PIL.Image.Image, np.ndarray)
 
     def _transform(
