Also convert splitval to tuple

fast_plotter/postproc/functions.py

@@ -201,14 +201,17 @@ def split(df, axis, keep_split_dim, return_meta=True):
     split the dataframe into a list of dataframes using a given binning
     dimensions
     """
-    if isinstance(axis, (list, tuple)):
-        axis = tuple(axis)
-    else:
-        axis = (axis, )
+    def to_tuple(obj):
+        if isinstance(obj, (list, tuple)):
+            return tuple(obj)
+        else:
+            return (obj, )
+    axis = to_tuple(axis)
     logger.info("Splitting on axis: '%s'", axis)
     out_dfs = []
     groups = df.groupby(level=axis, group_keys=keep_split_dim)
     for split_val, group in groups:
+        split_val = to_tuple(split_val)
         if not keep_split_dim:
             group.index = group.index.droplevel(axis)
         result = group.copy()