机器学习02-逻辑回归-python-优快云博客

本文链接：https://blog.youkuaiyun.com/tanliqing2010/article/details/84859525

本文深入探讨了逻辑回归的原理，包括sigmoid函数的作用及其损失函数的推导过程，并提供了从零开始的代码实现，最后对比了使用sklearn库进行逻辑回归的简便方法，适合初学者理解和实践。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

文章目录

1. 原理推导

2. 代码实现

def sigmod(x):
    return 1/(1 + np.exp(-x))
    
def compute_loss_theta(data , target , theta):
    m,n = data.shape
    h = sigmod(np.dot(data , theta))
    ## 这里是位置相乘，不是矩阵乘法
    cost1 = -1 *  np.sum(target * np.log(h) + (1-target) * np.log(1 - h)) / m
    dW = np.dot(data.T , (h - target) ) / m
    
    return dW , cost1

def train(X, y , alpha =0.01 , repeat=5000):
    lost = []
    theta = np.ones(X.shape[1]).reshape((-1, 1))
    for i in range(0 , repeat):
        dW , cost = compute_loss_theta(X , y , theta)
        # 梯度下降法
        theta = theta - alpha * dW
        lost.append(cost)
        if i %100 == 0:
            print 'i=%d , cost=%f' %(i,cost)
    
    return lost , theta

def predict(x_test ,  theta):
    return x_test.dot(theta)

3. sklearn实现

from sklearn.linear_model.logistic import LogisticRegression
import logic_regression 
from sklearn.cross_validation import train_test_split
X , y =  logic_regression.loadData()
clss = LogisticRegression()
X_train , X_test , y_train , y_test = train_test_split(X , y , train_size = 0.7)
clss.fit(X_train,y_train)
print clss.coef_
print clss.intercept_

print clss.score(X_test , y_test)