如何更改 Matplotlib 条形图上 x 轴标签上显示的单位

Question

我正在努力使收入的 x 轴上的刻度线将价值显示为百万分之一，而不是现在的一亿分之一。我似乎无法弄清楚如何做到这一点。我的代码和生成的条形图如下。

import numpy as np
import pandas as pd
import matplotlib.ticker as plticker
import matplotlib.pyplot as plt
import seaborn as sns
from IPython.display import display
from pandas import Series

%matplotlib inline


# Define Figure Size
fig, ax = plt.subplots(figsize=(25,25))

# Get the average vote_average for each genre
average_revenue = df2.groupby('release_year')['revenue_adj'].mean()

# Find positions of y ticks
y_pos = np.arange(len(df2['release_year'].unique()))
# Set up Bar Chart
ax.set_yticks(y_pos)
ax.set_yticklabels(sorted(df2['release_year'].unique()))
ax.set_xlabel('Revenue in Millions', fontsize=16)
ax.set_ylabel('Release Year', fontsize=16)
ax.set_title('Revenue by Release Year', fontsize=20)

# Set Size of X and Y labels
plt.rc('xtick', labelsize=14)
plt.rc('ytick', labelsize=14)

# Put Values next to Each Bar
for i, v in enumerate(average_revenue):
    a = v/1000000
    ax.text(v, i, ('$' + str(round(a,2)) + 'M'), color='blue')
    
ax.invert_yaxis()  # labels read top-to-bottom

# Draw Bar Chart
ax.barh(y_pos, average_revenue, align='center', color='green', ecolor='black')

Answer 1

目前，数据以个位显示，而不是数百万或数亿。注意图右侧的 1e8。您可以通过将输入除以一百万来绘制以百万为单位的值：

ax.barh(y_pos, average_revenue * 1e-6, ...)

或者，您可以这样调整 formatter on the x-axis if you prefer not to alter the data. For example, you could use a FuncFormatter：

from matplotlib.ticker import FuncFormatter

ax.xaxis.set_major_formatter(FuncFormatter(lambda x, pos: f'{x * 1e-6:0.1f}'))

如何更改 Matplotlib 条形图上 x 轴标签上显示的单位

How do I change the units shown on the x-axis labels on a Matplotlib bar chart

python

numpy

matplotlib

bar-chart

pandas