18

I would like to assign for x axis in matplotlib plot full date with time but with autoscale I could get only times or dates but not both. Following code:

import matplotlib.pyplot as plt
import pandas as pd

times = pd.date_range('2015-10-06', periods=500, freq='10min')

fig, ax = plt.subplots(1)
fig.autofmt_xdate()
plt.plot(times, range(times.size))
plt.show()

And on x axis I get only times without any dates so it's hard to distinct measurements.

I think that it's some option in matplotlib in matplotlib.dates.AutoDateFormatter but I couldn't find any one that could allow me to change that autoscale.

enter image description here

Anton Protopopov
  • 30,354
  • 12
  • 88
  • 93

3 Answers3

33

You can do this with a matplotlib.dates.DateFormatter, which takes a strftime format string as its argument. To get a day-month-year hour:minute format, you can use %d-%m-%y %H:%M:

import matplotlib.pyplot as plt
import pandas as pd
import matplotlib.dates as mdates

times = pd.date_range('2015-10-06', periods=500, freq='10min')

fig, ax = plt.subplots(1)
fig.autofmt_xdate()
plt.plot(times, range(times.size))

xfmt = mdates.DateFormatter('%d-%m-%y %H:%M')
ax.xaxis.set_major_formatter(xfmt)

plt.show()

enter image description here

tmdavison
  • 64,360
  • 12
  • 187
  • 165
  • 1
    If you define two y-axis for your plot, you may need to call `xaxis.set_major_formatter(xfmt)` on the other axis object. – Dom Apr 23 '21 at 15:50
  • how can we increase the number of dates (the frequency of showing the dates) on the x-axis? – Nihat Dec 03 '21 at 10:02
  • @Nihat you would want to change the `Locator`. For example, in this case we could use `ax.xaxis.set_major_locator(mdates.HourLocator(interval=6))` to have a tick every 6 hours. See [the docs here](https://matplotlib.org/stable/api/dates_api.html#date-tickers) for a list of different available locators. – tmdavison Dec 03 '21 at 14:29
1
plt.figure() 
plt.plot(...)
plt.gcf().autofmt_xdate() plt.show()
Yuchao Jiang
  • 3,522
  • 30
  • 23
0

This is how to plot date and time column when it is not your index column.
fig, ax = plt.subplots() ax.plot('docdate', 'count', data=newdf) fig.autofmt_xdate() plt.show()

FatiHe
  • 589
  • 7
  • 14