Logistic Regression

暂时自己没找数据跑

import numpy as np
import random
#import pandas as pd
def g_z(z):
    g = 1 / (1 + np.e**(-z))
    return g

def logis_function(k,x):
    x0 = np.ones((x.shape[0]))
    x = np.column_stack((x,x0))
    x_T = np.transpose(x)
    k = np.array(k)
    z = np.dot(k,x_T)
    pred_y = g_z(z)
    return pred_y

'''
这里的向量k最后一个元素是k0，即截距，这是因为在给定的x的最后一位都加
了元素1，以便在计算时，符合所有k，包括k0。
'''
def loss_func(k,x,y):
    y = np.array(y)
    pred_y = logis_function(k,x)
    single_loss = y * np.log(pred_y) + (1 - y) * np.log(1 - pred_y)
    loss = - np.sum(single_loss) / len(y)
    return loss

def gradient(k,x,y):
    x0 = np.ones((x.shape[0]))
    x1 = np.column_stack((x,x0))
    y = np.array(y)
    pred_y = logis_function(k,x)
    x_T = np.transpose(x1)
    grad_k = np.dot((pred_y - y), x_T) / len(y)
    return grad_k

def logistic_regression(x,y,epochs,lr,k):
    '''
    epochs 迭代次数
    lr 学习率
    k 参数向量
    '''
    best_k = None
    best_cost = np.inf
    k = np.array(k)
    for epoch in range(epochs):
        cost = loss_func(k,x,y)
        print("epoch: %d, k: %d, cost: %d" % (epoch,k,cost))
        if cost <= best_cost:
            best_cost = cost
            best_k = k
        grad_k = gradient(k,x,y)
        k = k - lr * grad_k

    return best_k, best_cost

def pred__y(k,x):
    pred_pro_y = logis_function(k,x)
    pred_pro_y[pred_pro_y > 0,5] = 1
    pred_pro_y[pred_pro_y <= 0.5] = 0
    return pred_pro_y