ENH Add colormap= keyword to DataFrame plots

Refactor color and colormap option handling

Add tests for colormap=

Add tests for colormap=

Add colormap documentation to visualization.rst

Add release notes

Author: Christopher Whelan

doc/source/release.rst

@@ -52,6 +52,8 @@ pandas 0.12
   - A ``filter`` method on grouped Series or DataFrames returns a subset of
     the original (:issue:`3680`, :issue:`919`)
   - Access to historical Google Finance data in pandas.io.data (:issue:`3814`)
+  - DataFrame plotting methods can sample column colors from a Matplotlib 
+    colormap via the ``colormap`` keyword. (:issue:`3860`)
 
 **Improvements to existing features**
 

doc/source/v0.12.0.txt

@@ -96,6 +96,12 @@ API changes
     and thus you should cast to an appropriate numeric dtype if you need to
     plot something.
 
+  - Add ``colormap`` keyword to DataFrame plotting methods. Accepts either a 
+    matplotlib colormap object (ie, matplotlib.cm.jet) or a string name of such 
+    an object (ie, 'jet'). The colormap is sampled to select the color for each 
+    column. Please see :ref:`visualization.colormaps` for more information.
+    (:issue:`3860`)
+
   - ``DataFrame.interpolate()`` is now deprecated. Please use
     ``DataFrame.fillna()`` and ``DataFrame.replace()`` instead. (:issue:`3582`,
     :issue:`3675`, :issue:`3676`)

doc/source/visualization.rst

@@ -531,3 +531,65 @@ be colored differently.
 
    @savefig radviz.png width=6in
    radviz(data, 'Name')
+
+.. _visualization.colormaps:
+
+Colormaps
+~~~~~~~~~
+
+A potential issue when plotting a large number of columns is that it can be difficult to distinguish some series due to repetition in the default colors. To remedy this, DataFrame plotting supports the use of the ``colormap=`` argument, which accepts either a Matplotlib `colormap <http://matplotlib.org/api/cm_api.html>`__ or a string that is a name of a colormap registered with Matplotlib. A visualization of the default matplotlib colormaps is available `here <http://wiki.scipy.org/Cookbook/Matplotlib/Show_colormaps>`__.
+
+As matplotlib does not directly support colormaps for line-based plots, the colors are selected based on an even spacing determined by the number of columns in the DataFrame. There is no consideration made for background color, so some colormaps will produce lines that are not easily visible.
+
+To use the jet colormap, we can simply pass ``'jet'`` to ``colormap=``
+
+.. ipython:: python
+
+   df = DataFrame(randn(1000, 10), index=ts.index)
+   df = df.cumsum()
+
+   plt.figure()
+
+   @savefig jet.png width=6in
+   df.plot(colormap='jet')
+
+or we can pass the colormap itself
+
+.. ipython:: python
+
+   from matplotlib import cm
+
+   plt.figure()
+
+   @savefig jet_cm.png width=6in
+   df.plot(colormap=cm.jet)
+
+Colormaps can also be used other plot types, like bar charts:
+
+.. ipython:: python
+
+   dd = DataFrame(randn(10, 10)).applymap(abs)
+   dd = dd.cumsum()
+
+   plt.figure()
+
+   @savefig greens.png width=6in
+   dd.plot(kind='bar', colormap='Greens')
+
+Parallel coordinates charts:
+
+.. ipython:: python
+
+   plt.figure()
+
+   @savefig parallel_gist_rainbow.png width=6in
+   parallel_coordinates(data, 'Name', colormap='gist_rainbow')
+
+Andrews curves charts:
+
+.. ipython:: python
+
+   plt.figure()
+
+   @savefig andrews_curve_winter.png width=6in
+   andrews_curves(data, 'Name', colormap='winter')

pandas/tests/test_graphics.py

@@ -103,6 +103,35 @@ def test_bar_colors(self):
             self.assert_(xp == rs)
 
         plt.close('all')
+
+        from matplotlib import cm
+
+        # Test str -> colormap functionality
+        ax = df.plot(kind='bar', colormap='jet')
+
+        rects = ax.patches
+
+        rgba_colors = map(cm.jet, np.linspace(0, 1, 5))
+        for i, rect in enumerate(rects[::5]):
+            xp = rgba_colors[i]
+            rs = rect.get_facecolor()
+            self.assert_(xp == rs)
+
+        plt.close('all')
+
+        # Test colormap functionality
+        ax = df.plot(kind='bar', colormap=cm.jet)
+
+        rects = ax.patches
+
+        rgba_colors = map(cm.jet, np.linspace(0, 1, 5))
+        for i, rect in enumerate(rects[::5]):
+            xp = rgba_colors[i]
+            rs = rect.get_facecolor()
+            self.assert_(xp == rs)
+
+        plt.close('all')
+
         df.ix[:, [0]].plot(kind='bar', color='DodgerBlue')
 
     @slow
@@ -600,6 +629,7 @@ def test_andrews_curves(self):
     def test_parallel_coordinates(self):
         from pandas import read_csv
         from pandas.tools.plotting import parallel_coordinates
+        from matplotlib import cm
         path = os.path.join(curpath(), 'data/iris.csv')
         df = read_csv(path)
         _check_plot_works(parallel_coordinates, df, 'Name')
@@ -611,6 +641,7 @@ def test_parallel_coordinates(self):
                           colors=('#556270', '#4ECDC4', '#C7F464'))
         _check_plot_works(parallel_coordinates, df, 'Name',
                           colors=['dodgerblue', 'aquamarine', 'seagreen'])
+        _check_plot_works(parallel_coordinates, df, 'Name', colormap=cm.jet)
 
         df = read_csv(
             path, header=None, skiprows=1, names=[1, 2, 4, 8, 'Name'])
@@ -622,9 +653,11 @@ def test_parallel_coordinates(self):
     def test_radviz(self):
         from pandas import read_csv
         from pandas.tools.plotting import radviz
+        from matplotlib import cm
         path = os.path.join(curpath(), 'data/iris.csv')
         df = read_csv(path)
         _check_plot_works(radviz, df, 'Name')
+        _check_plot_works(radviz, df, 'Name', colormap=cm.jet)
 
     @slow
     def test_plot_int_columns(self):
@@ -666,6 +699,7 @@ def test_line_colors(self):
         import matplotlib.pyplot as plt
         import sys
         from StringIO import StringIO
+        from matplotlib import cm
 
         custom_colors = 'rgcby'
 
@@ -691,6 +725,30 @@ def test_line_colors(self):
         finally:
             sys.stderr = tmp
 
+        plt.close('all')
+
+        ax = df.plot(colormap='jet')
+
+        rgba_colors = map(cm.jet, np.linspace(0, 1, len(df)))
+
+        lines = ax.get_lines()
+        for i, l in enumerate(lines):
+            xp = rgba_colors[i]
+            rs = l.get_color()
+            self.assert_(xp == rs)
+
+        plt.close('all')
+
+        ax = df.plot(colormap=cm.jet)
+
+        rgba_colors = map(cm.jet, np.linspace(0, 1, len(df)))
+
+        lines = ax.get_lines()
+        for i, l in enumerate(lines):
+            xp = rgba_colors[i]
+            rs = l.get_color()
+            self.assert_(xp == rs)
+
         # make color a list if plotting one column frame
         # handles cases like df.plot(color='DodgerBlue')
         plt.close('all')
@@ -862,6 +920,10 @@ def test_option_mpl_style(self):
         except ValueError:
             pass
 
+    def test_invalid_colormap(self):
+        df = DataFrame(np.random.randn(500, 2), columns=['A', 'B'])
+
+        self.assertRaises(ValueError, df.plot, colormap='invalid_colormap')
 
 def _check_plot_works(f, *args, **kwargs):
     import matplotlib.pyplot as plt