scikit-learn 中的随机梯度下降法

改进封装随机梯度下降法在线性回归中的使用：

def fit_sgd(self, X_train, y_train, n_iters=5, t0=5, t1=50):
    """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"
    assert n_iters >= 1

    def dJ_sgd(theta, X_b_i, y_i):
        return X_b_i * (X_b_i.dot(theta) - y_i) * 2.

    def sgd(X_b, y, initial_theta, n_iters, t0=5, t1=50):

        def learning_rate(t):
            return t0 / (t + t1)

        theta = initial_theta
        m = len(X_b)

        for cur_iter in range(n_iters):
            indexes = np.random.permutation(m)
            X_b_new = X_b[indexes]
            y_new = y[indexes]
            for i in range(m):
                gradient = dJ_sgd(theta, X_b_new[i], y_new[i])
                theta = theta - learning_rate(cur_iter * m + i) * gradient

        return theta

    X_b = np.hstack([np.ones((len(X_train), 1)), X_train])
    initial_theta = np.random.randn(X_b.shape[1])
    self._theta = sgd(X_b, y_train, initial_theta, n_iters, t0, t1)

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

    return self

#%% 使用我们自己的SGD

import numpy as np
import matplotlib.pyplot as plt

m = 100000

x = np.random.normal(size=m)
X = x.reshape(-1,1)
y = 4.* x + 3. + np.random.normal(0,3,size=m)

from machine_learning.playML.LinearRegression1 import LinearRegression

lin_reg = LinearRegression()
lin_reg.fit_sgd(X,y,n_iters=2)

结果为：

lin_reg.coef_
array([3.97761044])
lin_reg.intercept_
2.985958730191038

加载波士顿房价真实数据进行观察:

#%% 真实使用我们自己的SGD

from sklearn import datasets

boston = datasets.load_boston()
X = boston.data
y = boston.target

X = X[y < 50.0]
y = y[y < 50.0]

from machine_learning.playML.model_selection import train_test_split

X_train,y_train,X_test,y_test = train_test_split(X,y,seed=666)

#%% 归一化处理

from sklearn.preprocessing import StandardScaler

standardScaler = StandardScaler()
standardScaler.fit(X_train)
X_train_standard = standardScaler.transform(X_train)
X_test_standard = standardScaler.transform(X_test)

改变 n_iters 的值查看结果：

n_iters = 2

from machine_learning.playML.LinearRegression1 import LinearRegression

lin_reg1 = LinearRegression()
%time lin_reg1.fit_sgd(X_train_standard,y_train,n_iters=2)
lin_reg1.score(X_test_standard,y_test)

结果：

0.7857275413602651

n_iters = 50

0.8085607570556209

n_iters = 100

0.8129434245278827

再看看 sklearn 中的 SGD

from sklearn.linear_model import SGDRegressor   #只能解决线性问题
sgd_reg = SGDRegressor()
%time sgd_reg.fit(X_train_standard,y_train)
sgd_reg.score(X_test_standard,y_test)

结果：

Wall time: 3.98 ms
0.8126437519341116

添加参数：

sgd_reg = SGDRegressor(n_iter_no_change=100)  #无参数时，代表默认值 5 
%time sgd_reg.fit(X_train_standard,y_train)
sgd_reg.score(X_test_standard,y_test)

结果：

Wall time: 17 ms
0.8128008956699766