Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Upright mu in plot label: retaining original tick fonts

I have a problem which I thought would be more occurring. However, after scouring the internet for some time now I have not been able to find the solution to my problem. So here it goes:

For a plot, created using matplotlib.pyplot I want to incorporate the SI-unit micro meter into my xlabel. The unit micro meter however needs to be upright. After some fiddling around I achieved the desired xlabel.

The code that I have to generate this is:

import matplotlib
import matplotlib.pyplot as plt

matplotlib.rc('text', usetex = True)
params = {'text.latex.preamble': [r'\usepackage{siunitx}', r'\usepackage{cmbright}']}
plt.rcParams.update(params)


fig = plt.figure()
ax = fig.add_subplot(1,1,1)

ax.set_xlabel('$\si{\micro \meter}$', fontsize = 16)
ax.set_ylabel("hi", fontsize = 16)

plt.savefig("test.png")

The result is shown below: enter image description here The micro meter is exactly as I want it to be. The problem however is that the font of the x and y ticks is altered. This is because of:

matplotlib.rc('text', usetex = True)

How can I reset the font values to their original values? Or how can I make sure the fonts are not changed when introducing tex?

As a reference, the original values I am referring to look like this: enter image description here Besides trying to revert the fonts back to their original values I also tried different methods of incorporating micro meter into my xlabel. The problem that arises here is that I it remains italic or it has a bold type setting. The micro meter I am looking for is the one given in the first figure.

I hope someone can help me solve this problem.

Thanx in advance

like image 602
The Dude Avatar asked Mar 12 '14 10:03

The Dude


People also ask

Which command is used to print a text at a specified location on the plot area in Python?

text() is the main function we can use to add text to a plot. We start by adding the string “function” with fontsize=12 to the location (10, 20). By default, the location is given in data coordinates. The coordinate system can be changed using the transform parameter.

How do you plot a continuous graph in Python?

Plotting a continuous function plot with Seaborn # using seaborn import seaborn as sns import numpy as np # create data x = np. linspace(10, 100,10) y=np. power(x,3) # draw the graph sns. set_style('dark') fig,ax= plt.

How do I highlight points in Matplotlib?

Matplotlib has a function named annotate() to add text in a specific location in a plot. We need to specify annotate() function the text we want to annotate the plot with and the x and y co-ordinates for the location of the text.

Can NumPy be used to make plots in python?

For plotting graphs in Python, we will use the Matplotlib library. Matplotlib is used along with NumPy data to plot any type of graph. From matplotlib we use the specific function i.e. pyplot(), which is used to plot two-dimensional data.


2 Answers

What worked for me was not to usetex, but to use Unicode:

ax.set_xlabel(u'\u03bc')

sets the label as a single upright mu.

This requires the following settings when loading matplotlib:

import matplotlib
matplotlib.rcParams['mathtext.fontset'] = 'cm'
matplotlib.rc('font', family='serif', serif='CMU Serif')
import matplotlib.pyplot as plt

Here I'm using the "Computer Modern Unicode" font from Sourceforge (highly recommended if you'd like consistency with writing typeset in LaTeX and its default Computer Modern font).

But any unicode font with the mu glyph should work. Actually, the mu from CMU Serif is not as aesthetically nice as the mu from SIunitx, but it is correct.

Python needed to be restarted for that to take effect.

like image 144
Evgeni Sergeev Avatar answered Sep 29 '22 20:09

Evgeni Sergeev


I am also struggling with such a problem, i.e. getting the tick labels and axes labels to be consistent when text.usetex = True. The solution that I have managed to find it not ideal, but it works for the moment.

What you have to do is set the font family to "sans-serif" and also add a latex package that uses sans-serif math fonts (sfmath -- make sure it is in your tex path!)

import matplotlib
import matplotlib.pyplot as plt

matplotlib.rc('text', usetex = True)
matplotlib.rc('font', **{'family':"sans-serif"})
params = {'text.latex.preamble': [r'\usepackage{siunitx}', 
    r'\usepackage{sfmath}', r'\sisetup{detect-family = true}',
    r'\usepackage{amsmath}']}   
plt.rcParams.update(params)   

fig = plt.figure(figsize = (4,4))
ax = fig.add_subplot(1,1,1)
ax.set_xlabel('$\si{\um} detection$')
ax.set_ylabel(r"$\mu \boldsymbol{\mu}$")
plt.show()                              

In addition, I had to tell the siunitx package to detect the font family, and I also had to add some additional text to the x-label in order for the detection to actually work (you can remove this text later and the label will still work after that).

For me this results in: enter image description here More generally, I have the following my ~/.matplotlib/matploblibrc file, for serif fonts:

font.family    : serif 
text.latex.preamble    :   \usepackage{mathptmx}

and for sans-serif:

font.family : sans-serif 
text.latex.preamble :   \usepackage{sfmath}
like image 24
turnerm Avatar answered Sep 29 '22 21:09

turnerm