如何将多个Q-Q图合并为一个3×2的Matplotlib子图？

需求背景 #

原本使用Matplotlib创建了6个独立的Q-Q图，每个图都绘制了距离Q-Q线±2标准差的异常值线，现在需要将这些图合并为一个3×2布局的子图。

单个Q-Q图的实现代码（效果符合预期） #

f = plt.figure(1)
sm.qqplot(x1, line='r', fit=False, loc = np.mean(x1), scale = np.std(x1))
abline(slope1, intercept1 + 2*np.std(x1))
abline(slope1, intercept1 - 2*np.std(x1))
plt.title('STATS250')
f.show()

g = plt.figure(2)
sm.qqplot(x2, line='r', fit=False, loc = np.mean(x2), scale = np.std(x2))
abline(slope2, intercept2 + 2*np.std(x2))
abline(slope2, intercept2 - 2*np.std(x2))
plt.title('DATASCI306')
g.show()

h = plt.figure(3)
sm.qqplot(x3, line='r', fit=False, loc = np.mean(x3), scale = np.std(x3))
abline(slope3, intercept3 + 2*np.std(x3))
abline(slope3, intercept3 - 2*np.std(x3))
plt.title('MATH217')
h.show()

绘制异常值线的辅助函数 #

def abline(slope, intercept):
    """Plot a line from slope and intercept"""
    axes = plt.gca()
    x_vals = np.array(axes.get_xlim())
    y_vals = intercept + slope * x_vals
    plt.plot(x_vals, y_vals, '--')
    return

最初的错误尝试 #

尝试用subplots创建3×2子图并循环绘制时，所有异常值线都出现在最后一个子图中：

fig, axes = plt.subplots(nrows=3, ncols=2, figsize=(9,20))
ax= axes.flatten()
for i in range(6):
    sm.qqplot(x[i], line='r', fit=False, loc = np.mean(x[i]), scale = np.std(x[i]), ax=ax[i])
    abline(slope[i], intercept[i] + 2*np.std(x[i]))
    abline(slope[i], intercept[i] - 2*np.std(x[i]))
plt.show()

最终解决方法 #

修改辅助函数，传入子图的axis句柄，确保线条绘制到对应子图：

def abline2(slope, intercept, ax):
    """Plot a line from slope and intercept"""
    axes = ax
    x_vals = np.array(axes.get_xlim())
    y_vals = intercept + slope * x_vals
    axes.plot(x_vals, y_vals, '--')
    return

调用sm.qqplot时指定ax参数，同时在调用abline2时传入对应子图的axis句柄：

sm.qqplot(x[i], line='r', fit=False, loc = np.mean(x[i]), scale = np.std(x[i]), ax=ax[i])
abline2(slope[i], intercept[i] + 2*np.std(x[i]), ax=ax[i])
abline2(slope[i], intercept[i] - 2*np.std(x[i]), ax=ax[i])

内容的提问来源于stack exchange，提问作者Hambone_McKool