sklearn实现一元线性回归

from sklearn.linear_model import LinearRegression
import numpy as np
import matplotlib.pyplot as plt


# 最小二乘法
def compute_error(k, b, x, y):
    total_error = 0
    for i in range(0, len(x)):
        total_error += (y[i] - (k * x[i] + b)) ** 2
    return total_error / float(len(x)) / 2.0


# 载入数据
data = np.genfromtxt("data.csv", delimiter=",")

x_data = data[:, 0, np.newaxis]
y_data = data[:, 1, np.newaxis]

model = LinearRegression()
model.fit(x_data, y_data)

print("斜率是：", "截距是：", model.coef_,model.intercept_)
print("误差是：", compute_error(model.coef_, model.intercept_, x_data, y_data))
plt.plot(x_data, y_data, 'b.')
plt.plot(x_data, model.predict(x_data), 'r')
plt.show()

速度很快，不需要设置学习率，可以很快地找到很优的解