WIP mean

examples/01-Detailed-Examples/06-compute-statistics.py

@@ -55,3 +55,19 @@
 
 minimum = displacement.min(axis="set_ids")
 print(minimum)
+
+###############################################################################
+# Compute the average displacement value for each component at each time-step
+# ---------------------------------------------------------------------------
+
+mean = displacement.mean()
+print("Mean over mesh")
+print(mean)
+
+###############################################################################
+# Compute the average displacement for each node and component across time
+# ------------------------------------------------------------------------
+
+mean = displacement.mean(axis=1)
+print("Mean over time")
+print(mean)

src/ansys/dpf/post/dataframe.py

@@ -78,6 +78,7 @@ def __init__(
         self._last_display_max_colwidth = display_max_colwidth
 
         self._last_minmax: dict = {"axis": None, "min": None, "max": None}
+        self._last_mean: dict = {"axis": None, "mean": None}
 
     @property
     def columns(self) -> MultiIndex:
@@ -836,7 +837,7 @@ def animate(
             save_as=save_as, deform_by=deform_by, scale_factor=scale_factor, **kwargs
         )
 
-    def min(self, axis: Union[int, str, None] = 0) -> Union[DataFrame, float]:
+    def min(self, axis: Union[int, str, None] = 0) -> DataFrame:
         """Return the minimum value over the requested axis.
 
         Parameters
@@ -850,7 +851,7 @@ def min(self, axis: Union[int, str, None] = 0) -> Union[DataFrame, float]:
         self._query_min_max(axis)
         return self._last_minmax["min"]
 
-    def max(self, axis: Union[int, str, None] = 0) -> Union[DataFrame, float]:
+    def max(self, axis: Union[int, str, None] = 0) -> DataFrame:
         """Return the maximum value over the requested axis.
 
         Parameters
@@ -864,6 +865,20 @@ def max(self, axis: Union[int, str, None] = 0) -> Union[DataFrame, float]:
         self._query_min_max(axis)
         return self._last_minmax["max"]
 
+    def mean(self, axis: Union[int, str, None] = 0) -> DataFrame:
+        """Return the mean value over the requested axis.
+
+        Parameters
+        ----------
+        axis:
+            Axis to perform average across.
+            Defaults to the MeshIndex (0).
+            Can also be the SetIndex (1).
+
+        """
+        self._query_mean(axis)
+        return self._last_mean["mean"]
+
     def _query_min_max(self, axis: Union[int, str, None]) -> None:
         """Create a DPF workflow based on the query arguments for min/max."""
         # Translate None query to empty dict
@@ -932,7 +947,11 @@ def _query_min_max(self, axis: Union[int, str, None]) -> None:
 
             index = self.index
             columns = MultiIndex(
-                indexes=[c for c in self.columns if c != self.columns.set_ids]
+                indexes=[
+                    c
+                    for c in self.columns
+                    if c != getattr(self.columns, ref_labels.set_ids)
+                ]
             )
 
             min_fc = wf.get_output("min", dpf.types.fields_container)
@@ -949,3 +968,58 @@ def _query_min_max(self, axis: Union[int, str, None]) -> None:
             columns=columns,
         )
         self._last_minmax["axis"] = axis
+
+    def _query_mean(self, axis: Union[int, str, None]) -> None:
+        """Create a DPF workflow based on the query arguments for mean."""
+        # Translate None query to empty dict
+        if axis in [None, 0, self.index.mesh_index.name]:
+            axis = 0
+        elif axis in [1, ref_labels.set_ids]:
+            axis = 1
+        else:
+            raise ValueError(f"'{axis}' is not an available axis value.")
+        # If same query as last and last is not None, do not change
+        if self._last_mean["axis"] == axis and not self._last_mean["axis"] is None:
+            return
+        # If in need of an update, create the appropriate workflow
+        wf = dpf.Workflow(server=self._fc._server)
+        wf.progress_bar = False
+
+        # If over mesh
+        if axis == 0:
+            sum_op = dpf.operators.math.accumulate_fc(
+                fields_container=self._fc,
+                server=self._fc._server,
+            )
+
+            mean_fc = sum_op.outputs.fields_container()
+
+            index = MultiIndex(
+                indexes=[i for i in self.index if i != self.index.mesh_index]
+            )
+            columns = self.columns
+
+        # If over time
+        else:
+            average_over_label_op = dpf.operators.math.average_over_label_fc(
+                fields_container=self._fc,
+                server=self._fc._server,
+            )
+
+            mean_fc = average_over_label_op.outputs.field()
+
+            index = self.index
+            columns = MultiIndex(
+                indexes=[
+                    c
+                    for c in self.columns
+                    if c != getattr(self.columns, ref_labels.set_ids)
+                ]
+            )
+
+        self._last_mean["mean"] = DataFrame(
+            data=mean_fc,
+            index=index,
+            columns=columns,
+        )
+        self._last_mean["axis"] = axis