CS_ALearn/ailearning
1
0
Fork 0
mirror of https://github.com/apachecn/ailearning.git synced 2026-02-11 22:35:35 +08:00
Code Issues Projects Releases Wiki Activity

添加回归前两种的注释和测试说明

Browse Source
This commit is contained in:
chenyyx
2017-03-24 22:26:48 +08:00
parent 40543b634f
commit aa3cd3f648
1 changed files with 42 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files

83
src/python/08.Predictive numerical data regression/regression.py
View File

@@ -1,45 +1,46 @@
'''
Create by ApacheCN-xy
Create by ApacheCN-小瑶
Date from 2017-02-27
'''
from numpy import *
import matplotlib.pylab as plt
def loadDataSet(fileName): #解析以tab键分隔的文件中的浮点数
numFeat = len(open(fileName).readline().split('\t')) - 1 #获得一行有几个字段
def loadDataSet(fileName): #解析以tab键分隔的文件中的浮点数
numFeat = len(open(fileName).readline().split('\t')) - 1 #获得每一行的输入数据，最后一个代表真实值
dataMat = []; labelMat = []
fr = open(fileName)
for line in fr.readlines(): #读取每一行
for line in fr.readlines(): #读取每一行
lineArr =[]
curLine = line.strip().split('\t') #删除每一行的开头和结尾的tab
for i in range(numFeat):#从0到3，不包含3
lineArr.append(float(curLine[i]))#将数据添加到lineArr List中
dataMat.append(lineArr)
labelMat.append(float(curLine[-1]))
curLine = line.strip().split('\t') #删除一行中以tab分隔的数据前后的空白符号
for i in range(numFeat): #从0到2，不包括2
lineArr.append(float(curLine[i]))#将数据添加到lineArr List中，每一行数据测试数据组成一个行向量
dataMat.append(lineArr) #将测试数据的输入数据部分存储到dataMat矩阵中
labelMat.append(float(curLine[-1]))#将每一行的最后一个数据，即真实的目标变量存储到labelMat矩阵中
return dataMat,labelMat
def standRegres(xArr,yArr): #线性回归
xMat = mat(xArr); yMat = mat(yArr).T
xTx = xMat.T*xMat
if linalg.det(xTx) == 0.0:
def standRegres(xArr,yArr): #线性回归
xMat = mat(xArr); yMat = mat(yArr).T #mat()函数将xArr，yArr转换为矩阵
xTx = xMat.T*xMat #矩阵乘法的条件是左矩阵的列数等于右矩阵的行数
if linalg.det(xTx) == 0.0: #因为要用到xTx的逆矩阵，所以事先需要确定计算得到的xTx是否可逆，条件是矩阵的行列式不为0
print ("This matrix is singular, cannot do inverse")
return
ws = xTx.I * (xMat.T*yMat)
ws = xTx.I * (xMat.T*yMat) #书中的公式，求得w的最优解
return ws
def lwlr(testPoint,xArr,yArr,k=1.0):
def lwlr(testPoint,xArr,yArr,k=1.0): #局部加权线性回归
xMat = mat(xArr); yMat = mat(yArr).T
m = shape(xMat)[0]
weights = mat(eye((m)))
m = shape(xMat)[0] #获得xMat矩阵的行数
weights = mat(eye((m))) #eye()返回一个对角线元素为1，其他元素为0的二维数组，创建权重矩阵
for j in range(m): #下面两行创建权重矩阵
diffMat = testPoint - xMat[j,:] #
weights[j,j] = exp(diffMat*diffMat.T/(-2.0*k**2))
diffMat = testPoint - xMat[j,:] #遍历数据集，计算每个样本点对应的权重值
weights[j,j] = exp(diffMat*diffMat.T/(-2.0*k**2))#k控制衰减的速度
xTx = xMat.T * (weights * xMat)
if linalg.det(xTx) == 0.0:
print ("This matrix is singular, cannot do inverse")
return
ws = xTx.I * (xMat.T * (weights * yMat))
ws = xTx.I * (xMat.T * (weights * yMat)) #计算出回归系数的一个估计
return testPoint * ws
def lwlrTest(testArr,xArr,yArr,k=1.0): #循环所有的数据点，并将lwlr运用于所有的数据点
@@ -64,7 +65,7 @@ def ridgeRegres(xMat,yMat,lam=0.2): #岭回归
xTx = xMat.T*xMat
denom = xTx + eye(shape(xMat)[1])*lam
if linalg.det(denom) == 0.0:
print "This matrix is singular, cannot do inverse"
print ("This matrix is singular, cannot do inverse")
return
ws = denom.I * (xMat.T*yMat)
return ws
@@ -100,7 +101,7 @@ def stageWise(xArr,yArr,eps=0.01,numIt=100):
#returnMat = zeros((numIt,n)) #测试代码删除
ws = zeros((n,1)); wsTest = ws.copy(); wsMax = ws.copy()
for i in range(numIt):
print ws.T
print (ws.T)
lowestError = inf;
for j in range(n):
for sign in [-1,1]:
@@ -143,7 +144,8 @@ def stageWise(xArr,yArr,eps=0.01,numIt=100):
# i += 1
# currentRow = soup.findAll('table', r="%d" % i)
# fw.close()
'''
from time import sleep
import json
import urllib2
@@ -163,10 +165,10 @@ def searchForSet(retX, retY, setNum, yr, numPce, origPrc):
for item in listOfInv:
sellingPrice = item['price']
if sellingPrice > origPrc * 0.5:
print "%d\t%d\t%d\t%f\t%f" % (yr,numPce,newFlag,origPrc, sellingPrice)
print ("%d\t%d\t%d\t%f\t%f" % (yr,numPce,newFlag,origPrc, sellingPrice))
retX.append([yr, numPce, newFlag, origPrc])
retY.append(sellingPrice)
except: print 'problem with item %d' % i
except: print ('problem with item %d' % i)
def setDataCollect(retX, retY):
searchForSet(retX, retY, 8288, 2006, 800, 49.99)
@@ -210,8 +212,9 @@ def crossValidation(xArr,yArr,numVal=10):
xMat = mat(xArr); yMat=mat(yArr).T
meanX = mean(xMat,0); varX = var(xMat,0)
unReg = bestWeights/varX
print "the best model from Ridge Regression is:\n",unReg
print "with constant term: ",-1*sum(multiply(meanX,unReg)) + mean(yMat)
print ("the best model from Ridge Regression is:\n",unReg)
print ("with constant term: ",-1*sum(multiply(meanX,unReg)) + mean(yMat))
'''
@@ -221,32 +224,29 @@ def crossValidation(xArr,yArr,numVal=10):
#test for xianxinghuigui
def regression1():
xArr, yArr = loadDataSet("ex0.txt")
#test for standRegression
def regression1():
xArr, yArr = loadDataSet("../../../testData/Regression_data.txt")
xMat = mat(xArr)
yMat = mat(yArr)
ws = standRegres(xArr, yArr)
fig = plt.figure()
ax = fig.add_subplot(111)
ax.scatter(xMat[:, 1].flatten(), yMat.T[:, 0].flatten().A[0])
ax = fig.add_subplot(111) #add_subplot(349)函数的参数的意思是，将画布分成3行4列图像画在从左到右从上到下第9块
ax.scatter(xMat[:, 1].flatten(), yMat.T[:, 0].flatten().A[0]) #scatter 的x是xMat中的第二列，y是yMat的第一列
xCopy = xMat.copy()
xCopy.sort(0)
yHat = xCopy * ws
ax.plot(xCopy[:, 1], yHat)
plt.show()
if __name__ == "__main__":
regression1()
#test for jiaquanhuigui
def regression1():
xArr, yArr = loadDataSet("ex0.txt")
#test for LWLR
def regression2():
xArr, yArr = loadDataSet("../../../testData/Regression_data.txt")
yHat = lwlrTest(xArr, xArr, yArr, 0.003)
xMat = mat(xArr)
srtInd = xMat[:,1].argsort(0)
srtInd = xMat[:,1].argsort(0) #argsort()函数是将x中的元素从小到大排列，提取其对应的index(索引)，然后输出
xSort=xMat[srtInd][:,0,:]
fig = plt.figure()
ax = fig.add_subplot(111)
@@ -255,4 +255,5 @@ if __name__ == "__main__":
plt.show()
if __name__ == "__main__":
regression1()
#regression1()
#regression2()