Merge pull request #4722 from jorisvandenbossche/slow-plotting

PERF: faster plotting of timeseries with PeriodIndex

Author: jreback

doc/source/release.rst

@@ -47,7 +47,7 @@ pandas 0.13
   - ``get_dummies`` works with NaN (:issue:`4446`)
   - Added a test for ``read_clipboard()`` and ``to_clipboard()`` (:issue:`4282`)
   - Added bins argument to ``value_counts`` (:issue:`3945`), also sort and
-  ascending, now available in Series method as well as top-level function.
+    ascending, now available in Series method as well as top-level function.
   - Text parser now treats anything that reads like inf ("inf", "Inf", "-Inf",
     "iNf", etc.) to infinity. (:issue:`4220`, :issue:`4219`), affecting
     ``read_table``, ``read_csv``, etc.
@@ -74,6 +74,8 @@ pandas 0.13
   - Better/cleaned up exceptions in core/common, io/excel and core/format
     (:issue:`4721`, :issue:`3954`), as well as cleaned up test cases in
     tests/test_frame, tests/test_multilevel (:issue:`4732`).
+  - Performance improvement of timesesies plotting with PeriodIndex and added
+    test to vbench (:issue:`4705` and :issue:`4722`)
 
 **API Changes**
 

pandas/tests/test_graphics.py

@@ -60,6 +60,7 @@ def test_plot(self):
         _check_plot_works(self.ts.plot, style='.', logx=True)
         _check_plot_works(self.ts.plot, style='.', loglog=True)
         _check_plot_works(self.ts[:10].plot, kind='bar')
+        _check_plot_works(self.iseries.plot)
         _check_plot_works(self.series[:5].plot, kind='bar')
         _check_plot_works(self.series[:5].plot, kind='line')
         _check_plot_works(self.series[:5].plot, kind='barh')

pandas/tseries/converter.py

@@ -111,6 +111,8 @@ def convert(values, units, axis):
         if (isinstance(values, valid_types) or com.is_integer(values) or
                 com.is_float(values)):
             return get_datevalue(values, axis.freq)
+        if isinstance(values, PeriodIndex):
+            return values.asfreq(axis.freq).values
         if isinstance(values, Index):
             return values.map(lambda x: get_datevalue(x, axis.freq))
         if isinstance(values, (list, tuple, np.ndarray)):

vb_suite/plotting.py

@@ -0,0 +1,25 @@
+from vbench.benchmark import Benchmark
+from datetime import datetime
+
+common_setup = """from pandas_vb_common import *
+
+try:
+    from pandas import date_range
+except ImportError:
+    def date_range(start=None, end=None, periods=None, freq=None):
+        return DateRange(start, end, periods=periods, offset=freq)
+
+"""
+
+#-----------------------------------------------------------------------------
+# Timeseries plotting
+
+setup = common_setup + """
+N = 2000
+M = 5
+df = DataFrame(np.random.randn(N,M), index=date_range('1/1/1975', periods=N))
+"""
+
+plot_timeseries_period = Benchmark("df.plot()", setup=setup, 
+                                   name='plot_timeseries_period')
+

vb_suite/suite.py

@@ -17,6 +17,7 @@
            'miscellaneous',
            'panel_ctor',
            'parser',
+           'plotting',
            'reindex',
            'replace',
            'sparse',