多元线性回归实现梯度下降

笔记：


代码实现：

在线性回归模型中使用梯度下降法

import numpy as np
import matplotlib.pyplot as plt

np.random.seed(666)#数据随机，保持一致
x = 2 * np.random.random(size=100)#100个样本，每个样本有一个特征
y = x * 3. + 4. + np.random.normal(size=100)#均值为0，方差为1

X = x.reshape(-1, 1)#100行，一列数据

X[:20]

array([[ 1.40087424],
       [ 1.68837329],
       [ 1.35302867],
       [ 1.45571611],
       [ 1.90291591],
       [ 0.02540639],
       [ 0.8271754 ],
       [ 0.09762559],
       [ 0.19985712],
       [ 1.01613261],
       [ 0.40049508],
       [ 1.48830834],
       [ 0.38578401],
       [ 1.4016895 ],
       [ 0.58645621],
       [ 1.54895891],
       [ 0.01021768],
       [ 0.22571531],
       [ 0.22190734],
       [ 0.49533646]])

y[:20]

array([ 8.91412688,  8.89446981,  8.85921604,  9.04490343,  8.75831915,
        4.01914255,  6.84103696,  4.81582242,  3.68561238,  6.46344854,
        4.61756153,  8.45774339,  3.21438541,  7.98486624,  4.18885101,
        8.46060979,  4.29706975,  4.06803046,  3.58490782,  7.0558176 ])

plt.scatter(x, y)
plt.show()

使用梯度下降法训练

#定义损失函数
def J(theta, X_b, y):
    #异常处理，防止溢出
    try:
        return np.sum((y - X_b.dot(theta))**2) / len(X_b)#除以样本数（多少行）
    except:
        return float('inf')

#求导
def dJ(theta, X_b, y):
    res = np.empty(len(theta))#，开一个空间，求导之后结果的长度和x里的元素数目相同
    res[0] = np.sum(X_b.dot(theta) - y)#第0项
    for i in range(1, len(theta)):
        res[i] = (X_b.dot(theta) - y).dot(X_b[:,i])#第i个特征对应的向量即第i列
    return res * 2 / len(X_b)#乘以系数，2/m

#梯度下降过程
def gradient_descent(X_b, y, initial_theta, eta, n_iters = 1e4, epsilon=1e-8):

    theta = initial_theta
    cur_iter = 0

    while cur_iter < n_iters:
        gradient = dJ(theta, X_b, y)
        last_theta = theta
        theta = theta - eta * gradient
        if(abs(J(theta, X_b, y) - J(last_theta, X_b, y)) < epsilon):
            break

        cur_iter += 1

    return theta

X_b = np.hstack([np.ones((len(x), 1)), x.reshape(-1,1)])#为原先的x添加一列
initial_theta = np.zeros(X_b.shape[1])
eta = 0.01

theta = gradient_descent(X_b, y, initial_theta, eta)

theta

array([ 4.02145786,  3.00706277])

封装我们的线性回归算法

实现：

import numpy as np
from .metrics import r2_score

class LinearRegression:

    def __init__(self):
        """初始化Linear Regression模型"""
        self.coef_ = None
        self.intercept_ = None
        self._theta = None

    def fit_normal(self, X_train, y_train):
        """根据训练数据集X_train, y_train训练Linear Regression模型"""
        assert X_train.shape[0] == y_train.shape[0], \
            "the size of X_train must be equal to the size of y_train"

        X_b = np.hstack([np.ones((len(X_train), 1)), X_train])
        self._theta = np.linalg.inv(X_b.T.dot(X_b)).dot(X_b.T).dot(y_train)

        self.intercept_ = self._theta[0]
        self.coef_ = self._theta[1:]

        return self

    def fit_gd(self, X_train, y_train, eta=0.01, n_iters=1e4):
        """根据训练数据集X_train, y_train, 使用梯度下降法训练Linear Regression模型"""
        assert X_train.shape[0] == y_train.shape[0], \
            "the size of X_train must be equal to the size of y_train"

        def J(theta, X_b, y):
            try:
                return np.sum((y - X_b.dot(theta)) ** 2) / len(y)
            except:
                return float('inf')

        def dJ(theta, X_b, y):
            res = np.empty(len(theta))
            res[0] = np.sum(X_b.dot(theta) - y)
            for i in range(1, len(theta)):
                res[i] = (X_b.dot(theta) - y).dot(X_b[:, i])
            return res * 2 / len(X_b)

        def gradient_descent(X_b, y, initial_theta, eta, n_iters=1e4, epsilon=1e-8):

            theta = initial_theta
            cur_iter = 0

            while cur_iter < n_iters:
                gradient = dJ(theta, X_b, y)
                last_theta = theta
                theta = theta - eta * gradient
                if (abs(J(theta, X_b, y) - J(last_theta, X_b, y)) < epsilon):
                    break

                cur_iter += 1

            return theta

        X_b = np.hstack([np.ones((len(X_train), 1)), X_train])
        initial_theta = np.zeros(X_b.shape[1])
        self._theta = gradient_descent(X_b, y_train, initial_theta, eta, n_iters)

        self.intercept_ = self._theta[0]
        self.coef_ = self._theta[1:]

        return self

    def predict(self, X_predict):
        """给定待预测数据集X_predict，返回表示X_predict的结果向量"""
        assert self.intercept_ is not None and self.coef_ is not None, \
            "must fit before predict!"
        assert X_predict.shape[1] == len(self.coef_), \
            "the feature number of X_predict must be equal to X_train"

        X_b = np.hstack([np.ones((len(X_predict), 1)), X_predict])
        return X_b.dot(self._theta)

    def score(self, X_test, y_test):
        """根据测试数据集 X_test 和 y_test 确定当前模型的准确度"""

        y_predict = self.predict(X_test)
        return r2_score(y_test, y_predict)

    def __repr__(self):
        return "LinearRegression()"

from play.LinearRegression import LinearRegression

lin_reg = LinearRegression()
lin_reg.fit_gd(X, y)

LinearRegression()

lin_reg.coef_#系数

array([ 3.00706277])

lin_reg.intercept_#截距

4.021457858204859