ML之UliR：利用非線性回歸，梯度下降法(迭代十萬(wàn)次)求出學(xué)習(xí)參數(shù)θ，進(jìn)而求得Cost函數(shù)最優(yōu)值

輸出結(jié)果

更新……

代碼設(shè)計(jì)

import numpy as np  
import random       

def genData(numPoints,bias,variance):  
    x = np.zeros(shape=(numPoints,2)) 
    y = np.zeros(shape=(numPoints))  
    for i in range(0,numPoints):    
        x[i][0]=1                
        x[i][1]=i                  
        y[i]=(i+bias)+random.uniform(0,1)%variance 
    return x,y

def gradientDescent(x,y,theta,alpha,m,numIterations): 
    xTran = np.transpose(x)         
    for i in range(numIterations):
        hypothesis = np.dot(x,theta) 
        loss = hypothesis-y      
        cost = np.sum(loss**2)/(2*m) 
        gradient=np.dot(xTran,loss)/m
        theta = theta-alpha*gradient 
        print ("Iteration %d | cost :%f" %(i,cost))
    return theta

x,y = genData(100, 25, 10)  #100行，
print ("x:")
print (x)
print ("y:")
print (y)

m,n = np.shape(x)
n_y = np.shape(y)  
  
print("m:"+str(m)+" n:"+str(n)+" n_y:"+str(n_y))
  
numIterations = 100000    
alpha = 0.0005           
theta = np.ones(n)    
theta= gradientDescent(x, y, theta, alpha, m, numIterations)
print(theta)

相關(guān)文章
ML之UliR：利用非線性回歸，梯度下降法(迭代十萬(wàn)次)求出學(xué)習(xí)參數(shù)θ，進(jìn)而求得Cost函數(shù)最優(yōu)值