绘制模型残差

使用的 seaborn 组件： set_theme()，residplot()

import numpy as np
import seaborn as sns
sns.set_theme(style="whitegrid")

# Make an example dataset with y ~ x
rs = np.random.RandomState(7)
x = rs.normal(2, 1, 75)
y = 2 + 1.5 * x + rs.normal(0, 2, 75)

# Plot the residuals after fitting a linear model
sns.residplot(x=x, y=y, lowess=True, color="g")