user5315346
user5315346

Reputation:

Specifying values for my x-axis using the matplotlib.pyplot?

I am not able to specify values ​​for my x-axis , using the matplotlib.pyplot.

In some images the chart.xticks(years) solves the problem , but it seems that when the set of x-axis values ​​is too small , it uses default values like ​​[0,1,2,...,N].

A case that works: Image

A case that does not works: Image

My code:

import matplotlib.pyplot as chart
from matplotlib import lines

   # Settings
    chart.title(file_name)
    chart.xlabel('Years')
    chart.ylabel('Committers/Contributions')
    chart.ylim([0,highest_value + 100])
    chart.xlim(first_year,2017)

    # Values
    committer_line = chart.plot(committers_dict.keys(),committers_dict.values(),'r',label='Committer')
    contribution_line = chart.plot(contributions_dict.keys(),contributions_dict.values(),'b--',label='Contribution')
    years = list(range(first_year,2017))
    chart.xticks(years)

    # Legend
    chart.legend()

    # Show/Save
    chart.savefig(images_path + file_name.replace('.txt','-commiter-contribution.eps'), format='eps')
    chart.show()

Upvotes: 1

Views: 4765

Answers (2)

Suever
Suever

Reputation: 65430

If you look in the lower right hand corner, you'll see that matplotlib simply changed the display of the xtick labels to use scientific notation to save space. To change this you will want to alter the tick label Formatter.

To disable this conversion to scientific notation, you can tweak the default formatter:

ax.get_xaxis().get_major_formatter().set_useOffset(False)

This edits the useOffset parameter of your ScalarFormatter.

If you want to disable this behavior by default, you can edit the axes.formatter.useoffset rcparam.

Upvotes: 1

armatita
armatita

Reputation: 13465

Matplotlib is putting the correct numbers but in scientific notation because there's no more room in the x-axis to put more stuff. That said you can interact directly with the x-axis and state what strings do you want, in what positions, using a value for rotation. The following example (data was randomized):

import matplotlib.pyplot as chart
from matplotlib import lines
import random

title = 'Title'
first_year = 1960
last_year = 2017
x = [year for year in range(first_year,2017)]
y1 = [random.randint(0,10) for year in range(first_year,2017)]
y2 = [random.randint(0,10) for year in range(first_year,2017)]
highest_value = max(y1+y2)

# Settings
chart.title(title)
chart.xlabel('Years')
chart.ylabel('Committers/Contributions')
chart.ylim([0,highest_value + 0.05*highest_value])
chart.xlim(first_year,last_year)

# Values
committer_line = chart.plot(x,y1,'r',label='Committer')
contribution_line = chart.plot(x,y2,'b--',label='Contribution')
years = list(range(first_year,last_year))
years_str = [str(i) for i in range(first_year,last_year)]
chart.xticks(years,years_str,rotation=45)

# Legend
chart.legend()

# Show/Save
#chart.savefig(images_path + file_name.replace('.txt','-commiter-contribution.eps'), format='eps')
chart.show()

, results in:

lots of dates in x axis

, which is a bit cluttered so giving a step to you "Label Code" range:

years = list(range(first_year,last_year,5)) # It's 5!
years_str = [str(i) for i in range(first_year,last_year,5)] # It's 5!

, will ease the x axis information:

a few dates on the x axis

Upvotes: 1

Related Questions