Seaborn地块中的纯符号

Question

I started using seaborn and I just use pyplot.plot function, and whenever I use log scales the axis ticks are presented as 10 to some power. Before using seaborn, line

    ax.xaxis.set_major_formatter(FormatStrFormatter('%.0f'))

was fixing the problem, but now with seaborn it does not. Any ideas how I can have plain notation on my axes?

Below is an example of the code. I would like the x axis to be of the form: 0.1, 1, 10 , 100 .. and not as it is now.

import matplotlib.pyplot as plt
import seaborn as sns
sns.set_style('ticks',  {'font.family': 'sans-serif'})
sns.set_context("paper", font_scale=1.3)
plt.rc('text', usetex=True)


x_lim=[0.1,100]
y_lim=[1.2001, 2.2001]
f = plt.figure()
plt.xscale('log')
plt.ylim(y_lim)
plt.xlim(x_lim)
plt.hlines(1.5, *x_lim, color='grey')
plt.show()

Answer 1

The key thing is the placement of the formatter line. It has to be after ax.set_xscale('log') .

Rather than using %.0f which removes all numbers after the decimal place (leaving 0 rather than 0.1 ), use %g , which will round to 1 significant figure. I've extended your example to use a subplot so that the format can be set:

import matplotlib.pyplot as plt
import matplotlib.ticker as tck
import seaborn as sns

sns.set_style('ticks',  {'font.family': 'sans-serif'})
sns.set_context("paper", font_scale=1.3)
plt.rc('text', usetex=True)


x_lim=[0.1,100]
y_lim=[1.2001, 2.2001]

f = plt.figure()
ax = f.add_subplot(plt.subplot(1, 1, 1))
ax.set_xscale('log')
ax.set_ylim(y_lim)
ax.set_xlim(x_lim)
ax.hlines(1.5, *x_lim, color='grey')
ax.xaxis.set_major_formatter(tck.FormatStrFormatter('%g'))

plt.show()

This produces the image: