使用辅助 y 轴时共享 x 轴

Question

When creating a chart with a secondary y-axis, which is based on a pandas dataframe, I am struggling to share the x-axis. I am mostly lending from the solution here , where I like the EDIT of the accepted answer.

When recreating this solution using my own toy dataframe, I am facing the issue that my x-axis doesn't look too good.

df = pd.DataFrame(np.random.randint(0, 100, (20, 2)),
                  index=pd.date_range('20190101', periods=20),
                  columns=list('AB'))

fig, ax = plt.subplots()

ax.plot(df.index, df['A'], 'blue', label='Line A')

ax2 = ax.twinx()
ax2.plot(df.index, df['B'], 'red', label='Line B')

lines = ax.get_lines() + ax2.get_lines()

ax.legend(lines, [l.get_label() for l in lines],
          loc='upper left', frameon=False, fontsize=20)

Answer 1

One way to solve this is to rotate the x-tick labels:

df = pd.DataFrame(np.random.randint(0, 100, (20, 2)),
                  index=pd.date_range('20190101', periods=20),
                  columns=list('AB'))

fig, ax = plt.subplots()
plt.xticks(rotation=70)

ax.plot(df.index, df['A'], 'blue', label='Line A')

ax2 = ax.twinx()
ax2.plot(df.index, df['B'], 'red', label='Line B')

lines = ax.get_lines() + ax2.get_lines()



ax.legend(lines, [l.get_label() for l in lines],
          loc='upper left', frameon=False, fontsize=20)

Output:

Another option is just to increase the figure size or decrease the label font-size so that the labels no longer overlap.

fig, ax = plt.subplots(figsize=(15,10))

Output:

Answer 2

You can use pandas' builtin command, which provides some interesting tick labels:

fig, ax = plt.subplots()
df.plot(y=['A','B'], secondary_y='B', ax=ax)

Output: