diff --git a/docs/tutorials/custom_raster_dataset.ipynb b/docs/tutorials/custom_raster_dataset.ipynb
index 99a981f9d2..29d55c6fd1 100644
--- a/docs/tutorials/custom_raster_dataset.ipynb
+++ b/docs/tutorials/custom_raster_dataset.ipynb
@@ -332,6 +332,10 @@
     "\n",
     "If your data only contains model inputs (such as images), use `is_image = True`. If your data only contains ground truth model outputs (such as segmentation masks), use `is_image = False` instead.\n",
     "\n",
+    "Consequently, the sample returned by the dataset/data loader will use the \"image\" key if *is_image* is True, otherwise it will use the \"mask\" key.\n",
+    "\n",
+    "For datasets with both model inputs and outputs, the recommended approach is to use 2 `RasterDataset` instances and combine them using an `IntersectionDataset`. See L7 Irish, L8 Biome, and I/O Bench for examples of this in `torchgeo/datasets`.\n",
+    "\n",
     "### `dtype`\n",
     "\n",
     "Defaults to float32 for `is_image == True` and long for `is_image == False`. This is what you want for 99% of datasets, but can be overridden for tasks like pixel-wise regression (where the target mask should be float32).\n",
diff --git a/torchgeo/datasets/geo.py b/torchgeo/datasets/geo.py
index 8233480443..9dd886723b 100644
--- a/torchgeo/datasets/geo.py
+++ b/torchgeo/datasets/geo.py
@@ -361,8 +361,8 @@ class RasterDataset(GeoDataset):
     #: The sample returned by the dataset/data loader will use the "image" key if
     #: *is_image* is True, otherwise it will use the "mask" key.
     #:
-    #: For datasets with both model inputs and outputs, a custom
-    #: :func:`~RasterDataset.__getitem__` method must be implemented.
+    #: For datasets with both model inputs and outputs, the recommended approach is
+    #: to use 2 `RasterDataset` instances and combine them using an `IntersectionDataset`.
     is_image = True
 
     #: True if data is stored in a separate file for each band, else False.