8-1 用线性回归创建最佳拟合曲线
创建regression.py
# -*- coding: utf-8 -*-
"""
Created on Sat Oct 22 23:36:45 2016
@author: enniu
"""
from numpy import *
def loanDataSet(filename):
numFeat=len(open(filename).readline().split('\t'))-1
"""
#查看文件第一行内容
#cc=open(filename).readline().split('\t')
#for i in cc:
# print i,
"""
datamat=[];labelmat=[]
fr=open(filename)
for line in fr:
linearr=[]
curline=line.strip().split('\t')
for i in range(numFeat):
linearr.append(float(curline[i]))
#print curline
datamat.append(linearr)
labelmat.append(float(curline[-1]))
return datamat,labelmat
#datamat形如[[1.0, 0.067732], [1.0, 0.42781],。。。 labelmat形如[3.176513, 3.816464, 4.550095,
def standregres(xarr,yarr):
xmat=mat(xarr);ymat=mat(yarr).T #此处误写为:，导致报错can't assign to function call
xtx=xmat.T*xmat
if linalg.det(xtx)==0:
print '无法求逆矩阵'
ws=xtx.I*(xmat.T*ymat)
return ws
if __name__=='__main__':
a,b=loanDataSet('/Users/enniu/Desktop/jqxx/ex0.txt')
c=standregres(a,b)
print a
print b
print c
调用regression.py，并拟合回归曲线
from numpy import *
import regression as cc #import regression package
import matplotlib.pyplot as plt
"""
import sys
reload(sys)
sys.setdefaultencoding("utf-8")
"""
xarr,yarr=cc.loanDataSet('/Users/enniu/Desktop/jqxx/ex0.txt')
ws=cc.standregres(xarr,yarr)
xmat=mat(xarr)
ymat=mat(yarr)
yhat=xmat*ws #(200*2)(2*1)=(200,1)
#作图
fig=plt.figure()
ax=fig.add_subplot(111)
ax.scatter(xmat[:,1].flatten().A[0],ymat.T.flatten().A[0]) #做出散点图。flatten用于平整矩阵。A[0]用于获取array第一列
xcopy=xmat.copy()
xcopy.sort(0) #在原数据集上排序
yhat=xcopy*ws
ax.plot(xcopy[:,1],yhat)
plt.show() #must add
8-2 局部加权线性回归函数
# -*- coding: utf-8 -*-
"""
Created on Sat Oct 22 23:36:45 2016
@author: enniu
"""
from numpy import *
def loanDataSet(filename):
numFeat=len(open(filename).readline().split('\t'))-1
"""
#查看文件第一行内容
#cc=open(filename).readline().split('\t')
#for i in cc:
# print i,
"""
datamat=[];labelmat=[]
fr=open(filename)
for line in fr:
linearr=[]
curline=line.strip().split('\t')
for i in range(numFeat):
linearr.append(float(curline[i]))
#print curline
datamat.append(linearr)
labelmat.append(float(curline[-1]))
return datamat,labelmat
#datamat形如[[1.0, 0.067732], [1.0, 0.42781],。。。 labelmat形如[3.176513, 3.816464, 4.550095,
def standregres(xarr,yarr):
xmat=mat(xarr);ymat=mat(yarr).T #此处误写为:，导致报错can't assign to function call
xtx=xmat.T*xmat
if linalg.det(xtx)==0:
print '无法求逆矩阵'
ws=xtx.I*(xmat.T*ymat)
return ws #,xtx,xmat,ymat
def lwlr(testpoint,xarr,yarr,k=1.0):
xmat=mat(xarr);ymat=mat(yarr).T
m=shape(xmat)[0] #shape(xmat)=(200,1),m=200
weights=mat(eye(m))
for j in range(m):
diffmat=testpoint-xmat[j,:]
weights[j,j]=exp(diffmat*diffmat.T/(-2*k**2))
xtx=xmat.T*(weights*xmat)
if linalg.det(xtx)==0:
print "行列式为0，无法取逆"
return
ws=xtx.I*(xmat.T*(weights*ymat))
return testpoint*ws
def lwlrtest(testarr,xarr,yarr,k=1.0):
m=shape(testarr)[0]
yhat=zeros(m) #注意是括号
for i in range(m):
yhat[i]=lwlr(testarr[i],xarr,yarr,k)
return yhat
if __name__=='__main__':
a,b=loanDataSet('/Users/enniu/Desktop/jqxx/ex0.txt')
n=lwlr(1,a,b)
print n
调用regression.py，并拟合回归曲线
from numpy import *
import regression as cc
import matplotlib.pyplot as plt
xarr,yarr=cc.loanDataSet('/Users/enniu/Desktop/jqxx/ex0.txt')
ws=cc.standregres(xarr,yarr)
xmat=mat(xarr)
ymat=mat(yarr)
yhat=cc.lwlrtest(xarr,xarr,yarr,0.003) #type=matrix
strind=xmat[:,1].argsort(0)
xsort=xmat[strind][:,0,:]
fig=plt.figure()
ax=fig.add_subplot(111)
ax.scatter(xmat[:,1].flatten().A[0],ymat.T.flatten().A[0],s=2,c='red')
ax.plot(xsort[:,1],yhat[strind])
plt.show()
