From aa3cd3f648894d33a1ff917d48bb2f2247a1cc99 Mon Sep 17 00:00:00 2001 From: chenyyx Date: Fri, 24 Mar 2017 22:26:48 +0800 Subject: [PATCH] =?UTF-8?q?=E6=B7=BB=E5=8A=A0=E5=9B=9E=E5=BD=92=E5=89=8D?= =?UTF-8?q?=E4=B8=A4=E7=A7=8D=E7=9A=84=E6=B3=A8=E9=87=8A=E5=92=8C=E6=B5=8B?= =?UTF-8?q?=E8=AF=95=E8=AF=B4=E6=98=8E?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../regression.py | 83 ++++++++++--------- 1 file changed, 42 insertions(+), 41 deletions(-) diff --git a/src/python/08.Predictive numerical data regression/regression.py b/src/python/08.Predictive numerical data regression/regression.py index be857aa7..9e8a6ffb 100644 --- a/src/python/08.Predictive numerical data regression/regression.py +++ b/src/python/08.Predictive numerical data regression/regression.py @@ -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() \ No newline at end of file + #regression1() + #regression2() \ No newline at end of file