[英]Plot numpy datetime64 with matplotlib
I have two numpy arrays 1D, one is time of measurement in datetime64 format, for example:我有两个 numpy 数组 1D,一个是 datetime64 格式的测量时间,例如:
array([2011-11-15 01:08:11, 2011-11-16 02:08:04, ..., 2012-07-07 11:08:00], dtype=datetime64[us])
and other array of same length and dimension with integer data.和其他具有整数数据的相同长度和维度的数组。
I'd like to make a plot in matplotlib time vs data.我想在 matplotlib 时间与数据中绘制一个图。 If I put the data directly, this is what I get:
如果我直接放数据,这就是我得到的:
plot(timeSeries, data)
Is there a way to get time in more natural units?有没有办法以更自然的单位获得时间? For example in this case months/year would be fine.
例如,在这种情况下,月/年就可以了。
EDIT:编辑:
I have tried Gustav Larsson's suggestion however I get an error:我尝试过 Gustav Larsson 的建议,但出现错误:
Out[128]:
[<matplotlib.lines.Line2D at 0x419aad0>]
---------------------------------------------------------------------------
OverflowError Traceback (most recent call last)
/usr/lib/python2.7/dist-packages/IPython/zmq/pylab/backend_inline.pyc in show(close)
100 try:
101 for figure_manager in Gcf.get_all_fig_managers():
--> 102 send_figure(figure_manager.canvas.figure)
103 finally:
104 show._to_draw = []
/usr/lib/python2.7/dist-packages/IPython/zmq/pylab/backend_inline.pyc in send_figure(fig)
209 """
210 fmt = InlineBackend.instance().figure_format
--> 211 data = print_figure(fig, fmt)
212 # print_figure will return None if there's nothing to draw:
213 if data is None:
/usr/lib/python2.7/dist-packages/IPython/core/pylabtools.pyc in print_figure(fig, fmt)
102 try:
103 bytes_io = BytesIO()
--> 104 fig.canvas.print_figure(bytes_io, format=fmt, bbox_inches='tight')
105 data = bytes_io.getvalue()
106 finally:
/usr/lib/pymodules/python2.7/matplotlib/backend_bases.pyc in print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, **kwargs)
1981 orientation=orientation,
1982 dryrun=True,
-> 1983 **kwargs)
1984 renderer = self.figure._cachedRenderer
1985 bbox_inches = self.figure.get_tightbbox(renderer)
/usr/lib/pymodules/python2.7/matplotlib/backends/backend_agg.pyc in print_png(self, filename_or_obj, *args, **kwargs)
467
468 def print_png(self, filename_or_obj, *args, **kwargs):
--> 469 FigureCanvasAgg.draw(self)
470 renderer = self.get_renderer()
471 original_dpi = renderer.dpi
/usr/lib/pymodules/python2.7/matplotlib/backends/backend_agg.pyc in draw(self)
419
420 try:
--> 421 self.figure.draw(self.renderer)
422 finally:
423 RendererAgg.lock.release()
/usr/lib/pymodules/python2.7/matplotlib/artist.pyc in draw_wrapper(artist, renderer, *args, **kwargs)
53 def draw_wrapper(artist, renderer, *args, **kwargs):
54 before(artist, renderer)
---> 55 draw(artist, renderer, *args, **kwargs)
56 after(artist, renderer)
57
/usr/lib/pymodules/python2.7/matplotlib/figure.pyc in draw(self, renderer)
896 dsu.sort(key=itemgetter(0))
897 for zorder, a, func, args in dsu:
--> 898 func(*args)
899
900 renderer.close_group('figure')
/usr/lib/pymodules/python2.7/matplotlib/artist.pyc in draw_wrapper(artist, renderer, *args, **kwargs)
53 def draw_wrapper(artist, renderer, *args, **kwargs):
54 before(artist, renderer)
---> 55 draw(artist, renderer, *args, **kwargs)
56 after(artist, renderer)
57
/usr/lib/pymodules/python2.7/matplotlib/axes.pyc in draw(self, renderer, inframe)
1995
1996 for zorder, a in dsu:
-> 1997 a.draw(renderer)
1998
1999 renderer.close_group('axes')
/usr/lib/pymodules/python2.7/matplotlib/artist.pyc in draw_wrapper(artist, renderer, *args, **kwargs)
53 def draw_wrapper(artist, renderer, *args, **kwargs):
54 before(artist, renderer)
---> 55 draw(artist, renderer, *args, **kwargs)
56 after(artist, renderer)
57
/usr/lib/pymodules/python2.7/matplotlib/axis.pyc in draw(self, renderer, *args, **kwargs)
1039 renderer.open_group(__name__)
1040
-> 1041 ticks_to_draw = self._update_ticks(renderer)
1042 ticklabelBoxes, ticklabelBoxes2 = self._get_tick_bboxes(ticks_to_draw, renderer)
1043
/usr/lib/pymodules/python2.7/matplotlib/axis.pyc in _update_ticks(self, renderer)
929
930 interval = self.get_view_interval()
--> 931 tick_tups = [ t for t in self.iter_ticks()]
932 if self._smart_bounds:
933 # handle inverted limits
/usr/lib/pymodules/python2.7/matplotlib/axis.pyc in iter_ticks(self)
876 Iterate through all of the major and minor ticks.
877 """
--> 878 majorLocs = self.major.locator()
879 majorTicks = self.get_major_ticks(len(majorLocs))
880 self.major.formatter.set_locs(majorLocs)
/usr/lib/pymodules/python2.7/matplotlib/dates.pyc in __call__(self)
747 def __call__(self):
748 'Return the locations of the ticks'
--> 749 self.refresh()
750 return self._locator()
751
/usr/lib/pymodules/python2.7/matplotlib/dates.pyc in refresh(self)
756 def refresh(self):
757 'Refresh internal information based on current limits.'
--> 758 dmin, dmax = self.viewlim_to_dt()
759 self._locator = self.get_locator(dmin, dmax)
760
/usr/lib/pymodules/python2.7/matplotlib/dates.pyc in viewlim_to_dt(self)
528 def viewlim_to_dt(self):
529 vmin, vmax = self.axis.get_view_interval()
--> 530 return num2date(vmin, self.tz), num2date(vmax, self.tz)
531
532 def _get_unit(self):
/usr/lib/pymodules/python2.7/matplotlib/dates.pyc in num2date(x, tz)
287 """
288 if tz is None: tz = _get_rc_timezone()
--> 289 if not cbook.iterable(x): return _from_ordinalf(x, tz)
290 else: return [_from_ordinalf(val, tz) for val in x]
291
/usr/lib/pymodules/python2.7/matplotlib/dates.pyc in _from_ordinalf(x, tz)
201 if tz is None: tz = _get_rc_timezone()
202 ix = int(x)
--> 203 dt = datetime.datetime.fromordinal(ix)
204 remainder = float(x) - ix
205 hour, remainder = divmod(24*remainder, 1)
OverflowError: signed integer is greater than maximum
Could this be an bug?这可能是一个错误吗? Or am I missing something.
或者我错过了什么。 I also tried something simple:
我也尝试了一些简单的事情:
import matplotlib.pyplot as plt
import numpy as np
dates=np.array(["2011-11-13", "2011-11-14", "2011-11-15", "2011-11-16", "2011-11-19"], dtype='datetime64[us]')
data=np.array([1, 2, 3, 4, 5])
plt.plot_date(dates, data)
plt.show()
I still get this error:我仍然收到此错误:
OverflowError: signed integer is greater than maximum
I don't understand what am I doing wrong.我不明白我做错了什么。 ipython 0.13, matplotlib 1.1, Ubuntu 12.04 x64.
ipython 0.13、matplotlib 1.1、Ubuntu 12.04 x64。
FINAL EDIT:最终编辑:
It seems that matplotlib doesn't support dtype=datetime64
, so I needed to convert the timeSeries
to ordinary datetime.datetime
from datetime
.看来,matplotlib不支持
dtype=datetime64
,所以我需要的转换timeSeries
普通datetime.datetime
从datetime
。
from datetime import datetime
a=np.datetime64('2002-06-28').astype(datetime)
plot_date(a,2)
You might want to try this:你可能想试试这个:
plot_date(timeSeries, data)
By default, the x axis will be considered a date axis, and ya regular one.默认情况下,x 轴将被视为日期轴,并且是常规轴。 This can be customized.
这可以定制。
Matplotlib>=2.2 natively supports plotting datetime64 arrays. Matplotlib>=2.2 本身支持绘制 datetime64 数组。 See https://github.com/matplotlib/matplotlib/blob/master/doc/users/prev_whats_new/whats_new_2.2.rst#support-for-numpydatetime64 :
请参阅https://github.com/matplotlib/matplotlib/blob/master/doc/users/prev_whats_new/whats_new_2.2.rst#support-for-numpydatetime64 :
Matplotlib has supported datetime.datetime dates for a long time in matplotlib.dates.
Matplotlib 在 matplotlib.dates 中长期支持 datetime.datetime 日期。 We now support numpy.datetime64 dates as well.
我们现在也支持 numpy.datetime64 日期。 Anywhere that dateime.datetime could be used, numpy.datetime64 can be used.
任何可以使用 dateime.datetime 的地方,都可以使用 numpy.datetime64。 eg:
例如:
time = np.arange('2005-02-01', '2005-02-02', dtype='datetime64[h]') plt.plot(time)
I had a similar porblem.我有一个类似的问题。 Sometimes, the date axis plotted corectly my np.datetim64 array and at other times it did not with the same time array, giving some non-recognizible integer values on the date axis instead.
有时,日期轴正确地绘制了我的 np.datetim64 数组,而在其他时候,它没有使用相同的时间数组,而是在日期轴上给出一些无法识别的整数值。
The reason turned out to my having set ax.xscale('linear') after first having worked with a logarithmmic scale.原因是我在第一次使用对数刻度后设置了 ax.xscale('linear') 。 Removing the ax.xscale('linear') solved the problem.
删除 ax.xscale('linear') 解决了这个问题。 A linear axis is not a datetime axis, I learned.
我了解到,线性轴不是日期时间轴。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.