update docstring for predict_patchwise

Author: davidwilby

deepsensor/model/model.py

@@ -680,11 +680,14 @@ def predict_patchwise(
         progress_bar: int = 0,
         verbose: bool = False,
     ) -> Prediction:
-        """Predict on a regular grid or at off-grid locations.
+        """Predict using a tasks loaded using a sliding window patching strategy. Uses the `predict` method.
+
+        .. versionadded:: 0.4.3
+            :py:func:`predict_patchwise()` method.
 
         Args:
             tasks (List[Task] | Task):
-                List of tasks containing context data.
+                List of tasks containing context data. Tasks for patchwise prediction must be generated by a task loader using the "sliding" patching strategy.
             data_processor (:class:`~.data.processor.DataProcessor`):
                 Used for unnormalising the coordinates of the bounding boxes of patches.
             X_t (:class:`xarray.Dataset` | :class:`xarray.DataArray` | :class:`pandas.DataFrame` | :class:`pandas.Series` | :class:`pandas.Index` | :class:`numpy:numpy.ndarray`):
@@ -742,6 +745,9 @@ def predict_patchwise(
                 predictions.
 
         Raises:
+            AttributeError
+                If ``tasks`` are not generated using the "sliding" patching strategy of TaskLoader,
+                i.e. if they do not have a ``bbox`` attribute.
             ValueError
                 If ``X_t`` is not an xarray object and
                 ``resolution_factor`` is not 1 or ``ar_subsample_factor`` is