SideHistogram supports coloring weighted average

holoviews/plotting/bokeh/chart.py

@@ -300,7 +300,8 @@ def get_data(self, element, ranges=None, empty=None):
             data = dict(top=element.values, left=element.edges[:-1],
                         right=element.edges[1:])
 
-        dim = element.get_dimension(0)
+        color_dims = self.adjoined.traverse(lambda x: x.handles.get('color_dim'))
+        dim = color_dims[0] if color_dims else None
         cmapper = self._get_colormapper(dim, element, {}, {})
         if cmapper:
             data[dim.name] = [] if empty else element.dimension_values(dim)