逻辑回归实战：手写代码实现对马疝病数据集（horseColic）的分类预测

import pandas as pd
import numpy as np
from sklearn.linear_model import LogisticRegression

#加载数据
def load_data(path):
    data = pd.read_csv(path, sep='\t', names=[i for i in range(22)])
    data = np.array(data).tolist()
    x = []; y = []
    for i in range(len(data)):
        y.append(data[i][-1])
        del data[i][-1]
        x.append(data[i])

    x = np.array(x)
    y = np.array(y)

    return x, y

def sigmoid(z):
    return 1 / (1 + np.exp(-z))

def gradient(X, Y, w):
    res = []
    for k in range(22):
        sum = 0.0
        for i in range(len(Y)):
            sum += (-(Y[i] - np.dot(w.T, X[i])) * X[i, k])
        res.append(sum)

    res = np.array(res)
    return res


#基于梯度下降法
def logistics():
    train_x, train_y = load_data('horse_colic/horseColicTraining.txt')
    test_x, test_y = load_data('horse_colic/horseColicTest.txt')

    X = np.column_stack((np.ones((299, 1), float), train_x))
    Y = train_y
    Y = Y.reshape(299, 1)

    learning_rate = 0.00001
    step = 200   #迭代次数
    w = np.random.random((22, 1))
    while step > 0:
        res = gradient(X, Y, w)
        for i in range(len(w)):
            w[i] -= (learning_rate * res[i])
        step -= 1

    test_x = np.column_stack((np.ones((67, 1), float), test_x))  #最前面加上一列1
    sum1 = 0
    for i in range(len(test_y)):
        res = np.dot(w.T, test_x[i])
        res = sigmoid(res)
        if res > 0.5 and test_y[i] == 1:
            sum1 += 1
        elif res <= 0.5 and test_y[i] == 0:
            sum1 += 1
        else:
            continue

    print('手写正确率：', sum1 / len(test_y))


#基于sklearn包
def sklearn_logistics():
    train_x, train_y = load_data('horse_colic/horseColicTraining.txt')
    test_x, test_y = load_data('horse_colic/horseColicTest.txt')

    clf = LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,
                            intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,
                            penalty='l2', random_state=None, solver='liblinear', tol=0.0001,
                            verbose=0, warm_start=False)
    clf.fit(train_x, train_y)
    print('调包正确率：', clf.score(test_x, test_y))


if __name__ == '__main__':
    print('请稍等...')
    logistics()
    sklearn_logistics()