BUG: groupby.hist legend should use group keys

doc/source/whatsnew/v1.1.0.rst

@@ -236,6 +236,8 @@ Other enhancements
   and :class:`~pandas.io.stata.StataWriterUTF8` (:issue:`26599`).
 - :meth:`HDFStore.put` now accepts `track_times` parameter. Parameter is passed to ``create_table`` method of ``PyTables`` (:issue:`32682`).
 - Make :class:`pandas.core.window.Rolling` and :class:`pandas.core.window.Expanding` iterable（:issue:`11704`)
+- :meth:`DataFrame.hist`, :meth:`Series.hist`, :meth:`DataFrameGroupby.hist`, and :meth:`SeriesGroupby.hist` have gained the ``legend`` argument. Set to True to show a legend in the histogram. (:issue:`6279`)
+
 
 .. ---------------------------------------------------------------------------
 

pandas/plotting/_core.py

@@ -1,4 +1,5 @@
 import importlib
+from typing import TYPE_CHECKING, Optional, Sequence, Tuple, Union
 
 from pandas._config import get_option
 
@@ -9,19 +10,23 @@
 
 from pandas.core.base import PandasObject
 
+if TYPE_CHECKING:
+    from pandas import DataFrame
+
 
 def hist_series(
     self,
     by=None,
     ax=None,
-    grid=True,
-    xlabelsize=None,
-    xrot=None,
-    ylabelsize=None,
-    yrot=None,
-    figsize=None,
-    bins=10,
-    backend=None,
+    grid: bool = True,
+    xlabelsize: Optional[int] = None,
+    xrot: Optional[float] = None,
+    ylabelsize: Optional[int] = None,
+    yrot: Optional[float] = None,
+    figsize: Optional[Tuple[int, int]] = None,
+    bins: Union[int, Sequence[int]] = 10,
+    legend: bool = False,
+    backend: Optional[str] = None,
     **kwargs,
 ):
     """
@@ -50,6 +55,11 @@ def hist_series(
         bin edges are calculated and returned. If bins is a sequence, gives
         bin edges, including left edge of first bin and right edge of last
         bin. In this case, bins is returned unmodified.
+    legend : bool, default False
+        Whether to show the legend.
+
+        ..versionadded:: 1.1.0
+
     backend : str, default None
         Backend to use instead of the backend specified in the option
         ``plotting.backend``. For instance, 'matplotlib'. Alternatively, to
@@ -82,26 +92,28 @@ def hist_series(
         yrot=yrot,
         figsize=figsize,
         bins=bins,
+        legend=legend,
         **kwargs,
     )
 
 
 def hist_frame(
-    data,
-    column=None,
+    data: "DataFrame",
+    column: Optional[Union[str, Sequence[str]]] = None,
     by=None,
-    grid=True,
-    xlabelsize=None,
-    xrot=None,
-    ylabelsize=None,
-    yrot=None,
+    grid: bool = True,
+    xlabelsize: Optional[int] = None,
+    xrot: Optional[float] = None,
+    ylabelsize: Optional[int] = None,
+    yrot: Optional[float] = None,
     ax=None,
-    sharex=False,
-    sharey=False,
-    figsize=None,
-    layout=None,
-    bins=10,
-    backend=None,
+    sharex: bool = False,
+    sharey: bool = False,
+    figsize: Optional[Tuple[int, int]] = None,
+    layout: Optional[Tuple[int, int]] = None,
+    bins: Union[int, Sequence[int]] = 10,
+    legend: bool = False,
+    backend: Optional[str] = None,
     **kwargs,
 ):
     """
@@ -154,6 +166,11 @@ def hist_frame(
         bin edges are calculated and returned. If bins is a sequence, gives
         bin edges, including left edge of first bin and right edge of last
         bin. In this case, bins is returned unmodified.
+    legend : bool, default False
+        Whether to show the legend.
+
+        ..versionadded:: 1.1.0
+
     backend : str, default None
         Backend to use instead of the backend specified in the option
         ``plotting.backend``. For instance, 'matplotlib'. Alternatively, to
@@ -203,6 +220,7 @@ def hist_frame(
         sharey=sharey,
         figsize=figsize,
         layout=layout,
+        legend=legend,
         bins=bins,
         **kwargs,
     )

pandas/plotting/_matplotlib/hist.py

@@ -225,6 +225,7 @@ def _grouped_hist(
     xrot=None,
     ylabelsize=None,
     yrot=None,
+    legend=False,
     **kwargs,
 ):
     """
@@ -243,15 +244,27 @@ def _grouped_hist(
     sharey : bool, default False
     rot : int, default 90
     grid : bool, default True
+    legend: : bool, default False
     kwargs : dict, keyword arguments passed to matplotlib.Axes.hist
 
     Returns
     -------
     collection of Matplotlib Axes
     """
 
+    if legend and "label" not in kwargs:
+        if isinstance(data, ABCDataFrame):
+            if column is None:
+                kwargs["label"] = data.columns
+            else:
+                kwargs["label"] = column
+        else:
+            kwargs["label"] = data.name
+
     def plot_group(group, ax):
         ax.hist(group.dropna().values, bins=bins, **kwargs)
+        if legend:
+            ax.legend()
 
     if xrot is None:
         xrot = rot
@@ -290,6 +303,7 @@ def hist_series(
     yrot=None,
     figsize=None,
     bins=10,
+    legend: bool = False,
     **kwds,
 ):
     import matplotlib.pyplot as plt
@@ -308,8 +322,11 @@ def hist_series(
         elif ax.get_figure() != fig:
             raise AssertionError("passed axis not bound to passed figure")
         values = self.dropna().values
-
+        if legend and "label" not in kwds:
+            kwds["label"] = self.name
         ax.hist(values, bins=bins, **kwds)
+        if legend:
+            ax.legend()
         ax.grid(grid)
         axes = np.array([ax])
 
@@ -334,6 +351,7 @@ def hist_series(
             xrot=xrot,
             ylabelsize=ylabelsize,
             yrot=yrot,
+            legend=legend,
             **kwds,
         )
 
@@ -358,6 +376,7 @@ def hist_frame(
     figsize=None,
     layout=None,
     bins=10,
+    legend: bool = False,
     **kwds,
 ):
     if by is not None:
@@ -376,6 +395,7 @@ def hist_frame(
             xrot=xrot,
             ylabelsize=ylabelsize,
             yrot=yrot,
+            legend=legend,
             **kwds,
         )
         return axes
@@ -401,11 +421,17 @@ def hist_frame(
     )
     _axes = _flatten(axes)
 
+    can_set_label = "label" not in kwds
+
     for i, col in enumerate(data.columns):
         ax = _axes[i]
+        if legend and can_set_label:
+            kwds["label"] = col
         ax.hist(data[col].dropna().values, bins=bins, **kwds)
         ax.set_title(col)
         ax.grid(grid)
+        if legend:
+            ax.legend()
 
     _set_ticks_props(
         axes, xlabelsize=xlabelsize, xrot=xrot, ylabelsize=ylabelsize, yrot=yrot

pandas/tests/plotting/test_groupby.py

@@ -2,10 +2,11 @@
 
 
 import numpy as np
+import pytest
 
 import pandas.util._test_decorators as td
 
-from pandas import DataFrame, Series
+from pandas import DataFrame, Index, Series
 import pandas._testing as tm
 from pandas.tests.plotting.common import TestPlotBase
 
@@ -65,3 +66,48 @@ def test_plot_kwargs(self):
 
         res = df.groupby("z").plot.scatter(x="x", y="y")
         assert len(res["a"].collections) == 1
+
+    @pytest.mark.parametrize("column, expected_axes_num", [(None, 2), ("b", 1)])
+    @pytest.mark.parametrize("label", [None, "d"])
+    def test_groupby_hist_with_legend(self, column, expected_axes_num, label):
+        # GH 6279
+        # Histogram can have a legend
+        expected_layout = (1, expected_axes_num)
+        expected_labels = label or column or [["a"], ["b"]]
+
+        index = Index(15 * ["1"] + 15 * ["2"], name="c")
+        df = DataFrame(np.random.randn(30, 2), index=index, columns=["a", "b"])
+        g = df.groupby("c")
+
+        kwargs = {"legend": True, "column": column}
+        # We get warnings if kwargs contains "label": None
+        if label is not None:
+            kwargs["label"] = label
+
+        ret = g.hist(**kwargs)
+        for (_, axes) in ret.iteritems():
+            self._check_axes_shape(
+                axes, axes_num=expected_axes_num, layout=expected_layout
+            )
+            for ax, expected_label in zip(axes[0], expected_labels):
+                self._check_legend_labels(ax, expected_label)
+
+    @pytest.mark.parametrize(
+        "label, expected_label", [(None, ["1", "2"]), ("d", ["d", "d"])]
+    )
+    def test_groupby_hist_series_with_legend(self, label, expected_label):
+        # GH 6279
+        # Histogram can have a legend
+        index = Index(15 * ["1"] + 15 * ["2"], name="c")
+        df = DataFrame(np.random.randn(30, 2), index=index, columns=["a", "b"])
+        g = df.groupby("c")
+
+        kwargs = {"legend": True}
+        # We get warnings if kwargs contains "label": None
+        if label is not None:
+            kwargs["label"] = label
+
+        axes = g["a"].hist(**kwargs)
+        for (_, ax) in axes.iteritems():
+            self._check_axes_shape(ax, axes_num=1, layout=(1, 1))
+            self._check_legend_labels(ax, expected_label)

pandas/tests/plotting/test_hist_method.py

@@ -6,7 +6,7 @@
 
 import pandas.util._test_decorators as td
 
-from pandas import DataFrame, Series
+from pandas import DataFrame, Index, Series
 import pandas._testing as tm
 from pandas.tests.plotting.common import TestPlotBase, _check_plot_works
 
@@ -129,6 +129,28 @@ def test_plot_fails_when_ax_differs_from_figure(self):
         with pytest.raises(AssertionError):
             self.ts.hist(ax=ax1, figure=fig2)
 
+    @pytest.mark.parametrize(
+        "by, expected_axes_num, expected_layout", [(None, 1, (1, 1)), ("b", 2, (1, 2))]
+    )
+    @pytest.mark.parametrize("label, expected_label", [(None, "a"), ("c", "c")])
+    def test_hist_with_legend(
+        self, by, expected_axes_num, expected_layout, label, expected_label
+    ):
+        # GH 6279
+        # Histogram can have a legend
+        index = 15 * ["1"] + 15 * ["2"]
+        s = Series(np.random.randn(30), index=index, name="a")
+        s.index.name = "b"
+
+        kwargs = {"legend": True, "by": by}
+        # We get warnings if kwargs contains "label": None
+        if label is not None:
+            kwargs["label"] = label
+
+        axes = _check_plot_works(s.hist, **kwargs)
+        self._check_axes_shape(axes, axes_num=expected_axes_num, layout=expected_layout)
+        self._check_legend_labels(axes, expected_label)
+
 
 @td.skip_if_no_mpl
 class TestDataFramePlots(TestPlotBase):
@@ -293,6 +315,33 @@ def test_hist_column_order_unchanged(self, column, expected):
 
         assert result == expected
 
+    @pytest.mark.parametrize("by", [None, "c"])
+    @pytest.mark.parametrize("column", [None, "b"])
+    @pytest.mark.parametrize("label", [None, "d"])
+    def test_hist_with_legend(self, by, column, label):
+        # GH 6279
+        # Histogram can have a legend
+        expected_axes_num = 1 if by is None and column is not None else 2
+        expected_layout = (1, expected_axes_num)
+        expected_labels = label or column or ["a", "b"]
+        if by is not None:
+            expected_labels = [expected_labels] * 2
+
+        index = Index(15 * ["1"] + 15 * ["2"], name="c")
+        df = DataFrame(np.random.randn(30, 2), index=index, columns=["a", "b"])
+
+        kwargs = {"legend": True, "by": by, "column": column}
+        # We get warnings if kwargs contains "label": None
+        if label is not None:
+            kwargs["label"] = label
+
+        axes = _check_plot_works(df.hist, **kwargs)
+        self._check_axes_shape(axes, axes_num=expected_axes_num, layout=expected_layout)
+        if by is None and column is None and label is None:
+            axes = axes[0]
+        for expected_label, ax in zip(expected_labels, axes):
+            self._check_legend_labels(ax, expected_label)
+
 
 @td.skip_if_no_mpl
 class TestDataFrameGroupByPlots(TestPlotBase):