matplotlib y 轴标签随着字体大小的变化而消失

Question

I have the following code to plot the contents of a dataframe.

Using pandas and matplotlib:

thedata = {'2013':[0.0,0.0]
          ,'2014':[0.0,0.0]
          ,'2015':[0.0,0.0]
          ,'2016':[1,0.0]
          ,'2017':[0.0,0.0]
          ,'2018':[1,0.0]}

my_df = pd.DataFrame(thedata, index=['Green cars','Red cars'])

    plt.figure(figsize=(7,3))
    my_ax = plt.gca()
    my_ax.clear()
    my_ax.yaxis.set_major_locator(MaxNLocator(integer=True))

    my_df.transpose().plot(kind='bar'
                , stacked=True  
                , ax=my_ax
                ).grid(True,'major','y')

    my_ax.legend(loc=9, bbox_to_anchor=(0.5, -0.1), frameon=False, ncol=2, fontsize=12 )
    plt.title('All the cars', fontsize = 12 )
    my_ax.set_xticklabels(my_ax.get_xticklabels(),rotation='horizontal', fontsize=12)
#     my_ax.set_yticklabels(my_ax.get_yticklabels(),fontsize=12)

The last line is commented out in order to get the output displayed. I want to increase the font size of the y axis labels to be the same as the x, but when i un-comment that line and run it, the y axis labels just disappear and nothing is displayed there.

Why is this happening and how do I fix it?

Edit: - create the dataframe; pandas 0.23.0, matplotlib 2.2.2

Answer 1

The problem you face here is that the ticklabels are not actually defined before the figure is drawn.

import matplotlib.pyplot as plt
from matplotlib.ticker import MaxNLocator

fig, ax = plt.subplots()
ax.plot([1,3,4,2])
ax.xaxis.set_major_locator(MaxNLocator(integer=True))
ax.set_xticklabels(ax.get_xticklabels(),fontsize=12)
plt.show()

You may draw the figure, before accessing those labels

fig, ax = plt.subplots()
ax.plot([1,3,4,2])
ax.xaxis.set_major_locator(MaxNLocator(integer=True))

fig.canvas.draw()
ax.set_xticklabels(ax.get_xticklabels(),fontsize=12)
plt.show()

However, in order to change the fontsize of the ticklabels, one would rather use

ax.tick_params(axis="x", labelsize=12)

or set the property to the labels, without actually setting the labels' content. Eg via setp

plt.setp(ax.get_xticklabels(), fontsize=12)