ML之PLiR之LARS：利用LARS算法求解ElasticNet回归类型问题(实数值评分预测)

ML之PLiR之LARS：利用LARS算法求解ElasticNet回归类型问题(实数值评分预测)

---

设计思路

更新……

输出结果

['"alcohol"', '"volatile acidity"', '"sulphates"', '"total sulfur dioxide"', '"chlorides"', '"fixed acidity"', '"pH"', '"free sulfur dioxide"', '"citric acid"', '"residual sugar"', '"density"']

1、LARS

2、10-fold cross validation

Minimum Mean Square Error 0.5873018933136459
Index of Minimum Mean Square Error 311

实现代码

#initialize a vector of coefficients beta
beta = [0.0] * ncols

#initialize matrix of betas at each step
betaMat = []
betaMat.append(list(beta))

#number of steps to take
nSteps = 350
stepSize = 0.004
nzList = []

for i in range(nSteps):
    #calculate residuals
    residuals = [0.0] * nrows
    for j in range(nrows):
        labelsHat = sum([xNormalized[j][k] * beta[k] for k in range(ncols)])
        residuals[j] = labelNormalized[j] - labelsHat

    #calculate correlation between attribute columns from normalized wine and residual
    corr = [0.0] * ncols

    for j in range(ncols):
        corr[j] = sum([xNormalized[k][j] * residuals[k] for k in range(nrows)]) / nrows

    iStar = 0
    corrStar = corr[0]

    for j in range(1, (ncols)):
        if abs(corrStar) < abs(corr[j]):
            iStar = j; corrStar = corr[j]

    beta[iStar] += stepSize * corrStar / abs(corrStar)
    betaMat.append(list(beta))

    nzBeta = [index for index in range(ncols) if beta[index] != 0.0]
    for q in nzBeta:
        if (q in nzList) == False:
            nzList.append(q)

nameList = [names[nzList[i]] for i in range(len(nzList))]

print(nameList)
for i in range(ncols):
    #plot range of beta values for each attribute
    coefCurve = [betaMat[k][i] for k in range(nSteps)]
    xaxis = range(nSteps)
    plot.plot(xaxis, coefCurve)