How to get data in DateTime format from plot with Events in python?

Question

I have data in iPython DataFrame like this

    Date                        X           Y           Z
0   2015-11-30 20:23:05.556     281.764900  -43.895060  8.714666
1   2015-11-30 20:23:05.757     192.519990  -44.636436  1.720552
2   2015-11-30 20:23:05.958     149.030600  -45.098050  1.958352
3   2015-11-30 20:23:06.171     140.707600  -44.622448  1.510729
4   2015-11-30 20:23:06.366     139.154890  -45.154003  4.783974
5   2015-11-30 20:23:06.564     138.875140  -44.790306  2.266093

The dtypes of the columns are these:

Date    datetime64[ns]
X              float64
Y              float64
Z              float64
dtype: object

I set the button_press_event to my fig like this

def onclick(event):
    print 'button=%d, x=%d, y=%d, xdata=%s, ydata=%f'%(
        event.button, event.x, event.y, event.xdata, event.ydata)
cid = fig.canvas.mpl_connect('button_press_event', onclick)

and I want to get data from x axes in Date format or String format, but after click in graph I have just this

button=1, x=152, y=339, xdata=735932.856811, ydata=8.518828

How to set output from xdata in String?

Niko Fohr · Accepted Answer · 2022-02-01 16:21:37Z

5

Matplotlib's date support pre-dates numpy's datetime support (and predates pandas entirely). Therefore, it uses its own internal date convention.

To convert the "raw" number you're getting back to a datetime, use matplotlib.dates.num2date.

As an example similar to what you're doing:

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

import matplotlib.dates as mdates

def on_click(event):
    if event.inaxes is not None:
        print mdates.num2date(event.xdata)

t = pd.date_range('2015-11-01', '2016-01-06', freq='H')
y = np.random.normal(0, 1, t.size).cumsum()

df = pd.DataFrame({'Y':y}, index=t)

fig, ax = plt.subplots()
df.plot(ax=ax)
fig.canvas.mpl_connect('button_press_event', on_click)
plt.show()

edited Feb 1, 2022 at 16:21

Niko Fohr

35.4k12 gold badges113 silver badges117 bronze badges

answered Jan 11, 2016 at 16:28

Joe Kington

287k73 gold badges621 silver badges474 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Jan Bednar Over a year ago

So actually the solution was just put mdates.num2date(event.xdata) instead of event.xdata. Thank you very much!

s.paszko Over a year ago

45 min of search internet to find this answer! Thank you!

TimelyR Huang · Accepted Answer · 2022-05-24 19:37:35Z

1

The answer from @Joe Kington is not ideal for this situation. This answer is a working solution to this problem (returned a very large number in event.xdata in matplotlib click event when plotting with pandas.plot): https://stackoverflow.com/a/54036315/8564548 Pandas themself solve the solution with this link

pd.Period(ordinal=int(event.xdata), freq=freq)

where freq is the freq of your plot, could be something like '1s', '1min', '1H'... etc

edited May 24, 2022 at 19:37

answered May 24, 2022 at 19:35

TimelyR Huang

112 bronze badges

Collectives™ on Stack Overflow

How to get data in DateTime format from plot with Events in python?

2 Answers 2

2 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related