You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
or even pass datetime.datetime objects. Not sure how feasible this is. I think the trick is setting the converter field on the subplot's X axis according to:
In [22]: ax.xaxis.convert_units??
Type: instancemethod
Base Class: <type 'instancemethod'>
String Form:<bound method XAxis.convert_units of <matplotlib.axis.XAxis object at 0x3809310>>
Namespace: Interactive
File: /home/wesm/code/repos/matplotlib/lib/matplotlib/axis.py
Definition: ax.xaxis.convert_units(self, x)
Source:
def convert_units(self, x):
if self.converter is None:
self.converter = munits.registry.get_converter(x)
if self.converter is None:
#print 'convert_units returning identity: units=%s, converter=%s'%(self.units, self.converter)
return x
ret = self.converter.convert(x, self.units, self)
#print 'convert_units converting: axis=%s, units=%s, converter=%s, in=%s, out=%s'%(self, self.units, self.converter, x, ret)
return ret
The text was updated successfully, but these errors were encountered:
* origin/master:
DOC: string args to how in resample pandas-dev#1355
BLD: turning isreleased to False
DOC: release notes to closepandas-dev#1349
ENH: Cython nancorr speeds up DataFrame.corr with method='pearson' by > 100x
DOC: new parser functionality pandas-dev#1347
DOC: another pass at release notes
DOC: what's new
DOC: release notes, what's new
DOC: release notes and what's new
BUG: parser with multiple date col and multiple index col pandas-dev#1344
TST: additional tests for parsers and minor code cleanup
ENH: KdePlot with DataFrame pandas-dev#1342. TST: frame kde and kde with logy pandas-dev#1341
BUG: respect logy argument in KdePlot, closepandas-dev#1341
BUG: set_xlim for time series plots pandas-dev#1339
BUG: PeriodIndex.map tries to get super(DatetimIndex, self)
BUG: fixed doc bug that caused latex build to fail
DOC: cleaned up parser doc string to stop sphinx from complaining
ENH: better error msg for fillna() with invalid method
Ideally would like to be able to do:
plt.set_xlim(['1/1/2005', '1/1/2012'])
or even pass datetime.datetime objects. Not sure how feasible this is. I think the trick is setting the
converter
field on the subplot's X axis according to:The text was updated successfully, but these errors were encountered: