Editing the date formatting of x-axis tick labels in matplotlib

Question

I am looking to edit the formatting of the dates on the x-axis. The picture below shows how they appear on my bar graph by default. I would like to remove the repetition of 'Dec' and '2012' and just have the actual date numbers along the x-axis.

Any suggestions as to how I can do this?

enter image description here

The answer here worked for me: http://stackoverflow.com/questions/30133280/pandas-bar-plot-changes-date-format — Mahdi, Sep 21 '16 at 08:53

score 123 · Accepted Answer · answered Feb 18 '13 at 22:26

123

In short:

import matplotlib.dates as mdates
myFmt = mdates.DateFormatter('%d')
ax.xaxis.set_major_formatter(myFmt)

Many examples on the matplotlib website. The one I most commonly use is here

answered Feb 18 '13 at 22:26

Paul H

59,172
18
144
130

11

Is there an alternative method without using the `ax.xaxis` but using `plt.`? – 3kstc Jun 16 '17 at 05:45
2

@3kstc Maybe, but I can't imagine a situation where ambiguously relying on the state machine is preferable to the more explicit command. – Paul H Jun 16 '17 at 15:56
5

For a quick view of all possible datetime formats, use [strftime.org](http://strftime.org). – ImportanceOfBeingErnest Feb 08 '18 at 19:14
My dates are timezone aware and the DateFormatter converts times to UTC. I needed to add the timezone like so: mdates.DateFormatter('%H:%M', pytz.timezone('US/Eastern')). – Flint O'Brien Mar 17 '18 at 21:26
2

Note `ax` is defined in the answer by @Robino – khaverim Jan 24 '19 at 03:56
3

@3kstc If you want to use `plt` instead of `ax` you can do: `plt.gca().xaxis.set_major_formatter(myFmt)` – charelf Nov 11 '21 at 14:35
Thanks @charelf, lastly is there a way of changing the step/frequency of x ticks with using 'plt.'? – 3kstc Nov 12 '21 at 01:57
@3kstc `ax.xaxis.set_major_locator(....)` – Paul H Nov 12 '21 at 02:40
@PaulH I assume `plt.gca().xaxis.set_major_locator(...) ` – 3kstc Nov 12 '21 at 03:01
1

`plt.gca()` returns the Axes object, which you should define ahead of time – Paul H Nov 12 '21 at 06:57

score 48 · Answer 2 · edited Feb 08 '18 at 19:06

While the answer given by Paul H shows the essential part, it is not a complete example. On the other hand the matplotlib example seems rather complicated and does not show how to use days.

So for everyone in need here is a full working example:

from datetime import datetime
import matplotlib.pyplot as plt
from matplotlib.dates import DateFormatter

myDates = [datetime(2012,1,i+3) for i in range(10)]
myValues = [5,6,4,3,7,8,1,2,5,4]
fig, ax = plt.subplots()
ax.plot(myDates,myValues)

myFmt = DateFormatter("%d")
ax.xaxis.set_major_formatter(myFmt)

## Rotate date labels automatically
fig.autofmt_xdate()
plt.show()

score 3 · Answer 3 · answered Oct 24 '20 at 04:59

From the package matplotlib.dates as shown in this example the date format can be applied to the axis label and ticks for plot.

Below I have given an example for labeling axis ticks for multiplots

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

df = pd.read_csv('US_temp.csv')
plt.plot(df['Date'],df_f['MINT'],label='Min Temp.')
plt.plot(df['Date'],df_f['MAXT'],label='Max Temp.')
plt.legend()
####### Use the below functions #######
dtFmt = mdates.DateFormatter('%b') # define the formatting
plt.gca().xaxis.set_major_formatter(dtFmt) # apply the format to the desired axis
plt.show()

As simple as that

score 0 · Answer 4 · edited May 03 '22 at 09:58

This wokrs prfectly for me

import matplotlib.pyplot as plt
from matplotlib.ticker import (MultipleLocator, FormatStrFormatter,
                               AutoMinorLocator)
import matplotlib.dates as mdates

dtFmt = mdates.DateFormatter('%Y-%b') # define the formatting
plt.gca().xaxis.set_major_formatter(dtFmt) 
# show every 12th tick on x axes
plt.gca().xaxis.set_major_locator(mdates.MonthLocator(interval=1))
plt.xticks(rotation=90, fontweight='light',  fontsize='x-small',)

Editing the date formatting of x-axis tick labels in matplotlib

4 Answers4

Linked

Related