Add percent-based normalization in histplot (#2461)

* Add percent-based normalization in histplot

* Test bivariate normalization

(cherry picked from commit d0acb8c)

doc/releases/v0.12.0.txt

@@ -19,20 +19,6 @@ v0.12.0 (Unreleased)
 
 - |Enhancement| In :func:`histplot`, added `stat="percent"` as an option for normalization such that bar heights sum to 100 (:pr:`2461`).
 
-- |Enhancement| In :func:`histplot`, improved performance with large datasets and many groupings/facets (:pr:`2559`, :pr:`2570`).
-
-- |Enhancement| In :func:`kdeplot`, added the `warn_singular` parameter to allow silencing of the warning about data with zero variance (:pr:`2566`).
-
-- |Enhancement| In :class:`FacetGrid`, :class:`PairGrid`, and functions that use them, the matplotlib `figure.autolayout` parameter is disabled to avoid having the legend overlap the plot (:pr:`2571`).
-
-- |Enhancement| In :class:`FacetGrid` and functions that use it, the visibility of interior axis labels is now disabled, and exterior axis labels are no longer erased when adding additional layers (:pr:`2583`).
-
-- |Enhancement| |Fix| Improved integration with the matplotlib color cycle in most axes-level functions (:pr:`2449`).
-
-- |Feature| Added a ``refline`` method to :class:`FacetGrid` and :class:`JointGrid` for including horizontal and/or vertical reference lines using :meth:`matplotlib.axes.Axes.axhline`/:meth:`matplotlib.axes.Axes.axvline` (:pr:`2620`).
-
-- |API| |Enhancement| In :func:`lmplot`, added a new `facet_kws` parameter and deprecated the `sharex`, `sharey`, and `legend_out` parameters from the function signature; pass them in the `facet_kws` dictionary (:pr:`2576`).
-
 - |Fix| In :func:`lineplot, allowed the `dashes` keyword to set the style of a line without mapping a `style` variable (:pr:`2449`).
 
 - |Fix| In :func:`rugplot`, fixed a bug that prevented the use of datetime data (:pr:`2458`).

seaborn/_statistics.py

@@ -204,7 +204,7 @@ def __init__(
 
         Parameters
         ----------
-        stat : {"count", "frequency", "density", "probability"}
+        stat : {"count", "frequency", "density", "probability", "percent"}
             Aggregate statistic to compute in each bin.
 
             - ``count`` shows the number of observations
@@ -229,7 +229,8 @@ def __init__(
             If True, return the cumulative statistic.
 
         """
-        _check_argument("stat", ["count", "density", "probability", "frequency"], stat)
+        stat_choices = ["count", "frequency", "density", "probability", "percent"]
+        _check_argument("stat", stat_choices, stat)
 
         self.stat = stat
         self.bins = bins
@@ -337,6 +338,8 @@ def _eval_bivariate(self, x1, x2, weights):
 
         if self.stat == "probability":
             hist = hist.astype(float) / hist.sum()
+        elif self.stat == "percent":
+            hist = hist.astype(float) / hist.sum() * 100
         elif self.stat == "frequency":
             hist = hist.astype(float) / area
 
@@ -361,6 +364,8 @@ def _eval_univariate(self, x, weights):
 
         if self.stat == "probability":
             hist = hist.astype(float) / hist.sum()
+        elif self.stat == "percent":
+            hist = hist.astype(float) / hist.sum() * 100
         elif self.stat == "frequency":
             hist = hist.astype(float) / np.diff(bin_edges)
 

seaborn/tests/test_distributions.py

@@ -1258,6 +1258,12 @@ def test_probability_stat_unique_norm(self, long_df):
             bar_heights = [b.get_height() for b in bars]
             assert sum(bar_heights) == pytest.approx(1)
 
+    def test_percent_stat(self, flat_series):
+
+        ax = histplot(flat_series, stat="percent")
+        bar_heights = [b.get_height() for b in ax.patches]
+        assert sum(bar_heights) == 100
+
     def test_common_bins(self, long_df):
 
         n = 10
@@ -1796,6 +1802,17 @@ def test_mesh_unique_norm(self, long_df):
             density, (x_edges, y_edges) = sub_hist(sub_df["x"], sub_df["y"])
             assert_array_equal(mesh_data.data, density.T.flat)
 
+    @pytest.mark.parametrize("stat", ["probability", "percent"])
+    def test_mesh_normalization(self, long_df, stat):
+
+        ax = histplot(
+            long_df, x="x", y="y", stat=stat,
+        )
+
+        mesh_data = ax.collections[0].get_array()
+        expected_sum = {"probability": 1, "percent": 100}[stat]
+        assert mesh_data.data.sum() == expected_sum
+
     def test_mesh_colors(self, long_df):
 
         color = "r"