如何让Matplotlib多个子图的X轴刻度垂直对齐？

从DataFrame的特定行绘制3个子图时出现异常：第三个子图的X轴刻度间距与前两个不同，尽管绘图起始位置看似正确。推测原因是FE 57和FE 59的索引接近0，而FE 66的索引从3600开始。需要让所有子图从0开始保留空白，使X轴刻度垂直对齐。

现有代码 #

fig, axs = plt.subplots(3, 1)
ser = plotdf[(plotdf['CAN_address'] == 'FE 57') & (plotdf['CAN_cmd'] == 62)]['CAN_value_translated']
axs[0].plot(ser.index, ser.values)
axs[0].set_title('FE 57')

ser = plotdf[(plotdf['CAN_address'] == 'FE 59') & (plotdf['CAN_cmd'] == 62)]['CAN_value_translated']
axs[1].plot(ser.index, ser.values)
axs[1].set_title('FE 59')

ser = plotdf[(plotdf['CAN_address'] == 'FE 66') & (plotdf['CAN_cmd'] == 62)]['CAN_value_translated']
axs[2].plot(ser.index, ser.values)
axs[2].set_title('FE 66')

plt.show()

异常效果 #

方法一：手动统一X轴范围 #

先提取所有目标数据的最大索引值，再给每个子图设置相同的X轴范围（从0到该最大值）：

# 获取三类数据的最大索引
max_index = max(
    plotdf[(plotdf['CAN_address'] == 'FE 57') & (plotdf['CAN_cmd'] == 62)].index.max(),
    plotdf[(plotdf['CAN_address'] == 'FE 59') & (plotdf['CAN_cmd'] == 62)].index.max(),
    plotdf[(plotdf['CAN_address'] == 'FE 66') & (plotdf['CAN_cmd'] == 62)].index.max()
)

fig, axs = plt.subplots(3, 1)
# 绘制FE 57
ser = plotdf[(plotdf['CAN_address'] == 'FE 57') & (plotdf['CAN_cmd'] == 62)]['CAN_value_translated']
axs[0].plot(ser.index, ser.values)
axs[0].set_title('FE 57')
axs[0].set_xlim(0, max_index)

# 绘制FE 59
ser = plotdf[(plotdf['CAN_address'] == 'FE 59') & (plotdf['CAN_cmd'] == 62)]['CAN_value_translated']
axs[1].plot(ser.index, ser.values)
axs[1].set_title('FE 59')
axs[1].set_xlim(0, max_index)

# 绘制FE 66
ser = plotdf[(plotdf['CAN_address'] == 'FE 66') & (plotdf['CAN_cmd'] == 62)]['CAN_value_translated']
axs[2].plot(ser.index, ser.values)
axs[2].set_title('FE 66')
axs[2].set_xlim(0, max_index)

plt.tight_layout()
plt.show()

方法二：共享X轴自动对齐 #

创建子图时设置sharex=True，让所有子图共用同一X轴，再强制X轴从0开始：

fig, axs = plt.subplots(3, 1, sharex=True)
# 绘制FE 57
ser = plotdf[(plotdf['CAN_address'] == 'FE 57') & (plotdf['CAN_cmd'] == 62)]['CAN_value_translated']
axs[0].plot(ser.index, ser.values)
axs[0].set_title('FE 57')

# 绘制FE 59
ser = plotdf[(plotdf['CAN_address'] == 'FE 59') & (plotdf['CAN_cmd'] == 62)]['CAN_value_translated']
axs[1].plot(ser.index, ser.values)
axs[1].set_title('FE 59')

# 绘制FE 66
ser = plotdf[(plotdf['CAN_address'] == 'FE 66') & (plotdf['CAN_cmd'] == 62)]['CAN_value_translated']
axs[2].plot(ser.index, ser.values)
axs[2].set_title('FE 66')

# 强制X轴起始为0
plt.xlim(left=0)
plt.tight_layout()
plt.show()

说明：sharex=True会自动同步所有子图的X轴刻度和范围，配合plt.xlim(left=0)能让第三个子图在0到3600区间显示空白，实现三个子图X轴刻度完全垂直对齐。

内容的提问来源于stack exchange，提问作者Anjo