梯度下降实现简单实现 gradient descent

#coding=utf-8

import numpy as np
import random
from numpy import genfromtxt
import sys

def getData(dataSet):
    m, n = np.shape(dataSet)
    print 'np.shape(dataSet):',np.shape(dataSet)
    trainData = np.ones((m, n))
    trainData[:,:-1] = dataSet[:,:-1]
    trainLabel = dataSet[:,-1]
    return trainData, trainLabel

def batchGradientDescent(x, y, theta, alpha, m, maxIterations):
    xTrains = x.transpose()
    print 'xTrains:',xTrains
    print 'x:',x
    print 'y:',y
    for i in range(0, maxIterations):
        print i,'=================='
        print 'theta:',theta
        hypothesis = np.dot(x, theta)
        print 'hypothesis:',hypothesis
        loss = hypothesis - y
        print 'loss:',loss
        # print loss
        gradient = np.dot(xTrains, loss) / m
        theta = theta - alpha * gradient
    return theta

def predict(x, theta):
    m, n = np.shape(x)
    xTest = np.ones((m, n+1))
    print 'xTest:',xTest
    xTest[:, :-1] = x
    print 'xTest:',xTest
    yP = np.dot(xTest, theta)
    return yP
if __name__ == "__main__":
    dataPath = "house.csv"
    dataSet = genfromtxt(dataPath, delimiter=',')
    print dataSet
    trainData, trainLabel = getData(dataSet)
    print 'trainData:',trainData
    print 'trainLabel:',trainLabel
    m, n = np.shape(trainData)
    print 'np.shape(trainData):',np.shape(trainData)
    theta = np.ones(n)
    print 'theta:',theta
    alpha = 0.1
    maxIteration = 5
    theta = batchGradientDescent(trainData, trainLabel, theta, alpha, m, maxIteration)
    print '========theta',theta
    x = np.array([[3.1, 5.5], [3.3, 5.9], [3.5, 6.3], [3.7, 6.7], [3.9, 7.1]])
    print predict(x, theta)

'''cat house.csv
1.1,1.5,1.0
2.1,2.3,0.0'''