logistic----------------------------python implement

#coding:utf-8

from numpy import *
import pandas as pd



theta=mat([0,0,0])   #初始theta值

alpha=0.001   #step

#datasets
file_name=("D:\\python_text\\logist.csv")

#加载数据集
def loadDataSet(file_name):
	data_x = pd.read_csv(file_name,usecols=[0,1])  #只读取前四列，最后一列是因变量
	data_y = pd.read_csv(file_name,usecols=[2]) #打印最后一行作为因变量
	X_array=mat(data_x)  #将csv中的数据读成一个矩阵,且第一行的标签也没有打印，有点神奇
	y_array=mat(data_y)
	_x0=ones(X_array.shape[0])
	_X_array=c_[_x0,X_array]
	return _X_array,y_array

#定义sigmoid函数
def sigmoid(list_x):
	return 1.0/(1+exp(-list_x))


def J(sig_z,y):
	return -(y.T*(log(sig_z.T)).T + (1-y.T)*(log(1-sig_z.T)).T)    #Loss function


#定义求解最佳回归系数
def gradAscent(dataMatIn,classLabels,theta):
	while (True):
		H_1=sigmoid(dataMatIn*(theta.T))
		trive_1=J(H_1, classLabels)  #calculate J(theta)
		theta=theta-alpha*((H_1-classLabels).T*data)      #update theta
		H_2=sigmoid(dataMatIn*(theta.T))	
		trive_2=J(H_2, classLabels)  #calculate  update J(theta)
		if(abs(trive_1-trive_2)<0.001):
			return theta

data,label=loadDataSet(file_name)
T_value=gradAscent(data, label,theta)


#画出决策边界

def plotBestFit(wei,dataMatrix,labelMat):
	import matplotlib.pyplot as plt
	weights = wei.getA()     #将矩阵wei转化为list
	dataArr = array(dataMatrix)  #将矩阵转化为数组
	n = shape(dataMatrix)[0]
	xcord1 = [];ycord1=[]
	xcord2 = [];ycord2=[]
 
	for i in range(n):
		if int(labelMat[i])==1:
			xcord1.append(dataArr[i,1])
			ycord1.append(dataArr[i,2])
		else:
			xcord2.append(dataArr[i,1])
			ycord2.append(dataArr[i,2])
		
	fig = plt.figure()
	ax = fig.add_subplot(111)
	ax.scatter(xcord1,ycord1,s=30,c='red', marker='s')
	ax.scatter(xcord2,ycord2,s=30,c="green")
	x = arange(-3.0,3.0,0.1)
	y = (-weights[0][0]-weights[0][1] * x)/weights[0][2]
	ax.plot(x,y)
	plt.xlabel("x1")     #X轴的标签
	plt.ylabel("x2")     #Y轴的标签
	plt.show()


plotBestFit(T_value, data, label)

我的loss是用最大似然函数