ML之回归预测：利用FSR/RiR/BasisExpand/ Lasso/DT/RF/GB算法对红酒品质wine数据集实现红酒口感评分预测(实数值评分预测)

ML之回归预测：利用FSR/RiR/BasisExpand/ Lasso/DT/RF/GB算法对红酒品质wine数据集实现红酒口感评分预测(实数值评分预测)

 

 

目录

输出结果

设计思路

T1、FSR(前向逐步回归)

T2、RiR(岭回归)

T3、基扩展BasisExpand

核心代码

---

 

 

 

输出结果

Index(['fixed acidity', 'volatile acidity', 'citric acid', 'residual sugar',
       'chlorides', 'free sulfur dioxide', 'total sulfur dioxide', 'density',
       'pH', 'sulphates', 'alcohol', 'quality'],
      dtype='object')

 

 

 

 

设计思路

T1、FSR(前向逐步回归)

Out of sample error versus attribute set size
[0.7234259255116278, 0.6860993152837196, 0.6734365033420278, 0.6677033213897796, 0.6622558568522273, 0.6590004754154626, 0.6572717206143076, 0.6570905806207697, 0.6569993096446138, 0.6575818940043473, 0.657390986901134]

Best attribute indices
[10, 1, 9, 4, 6, 8, 5, 3, 2, 7, 0]

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

 

T2、RiR(岭回归)

RMS Error             alpha
0.6595788176342458 1.0
0.6578610918808593 0.1
0.6576172144640243 0.010000000000000002
0.6575216482641756 0.0010000000000000002
0.6574190680109294 0.00010000000000000002
0.6573941628851253 1.0000000000000003e-05
0.6573913087155858 1.0000000000000004e-06

T3、基扩展BasisExpand

 

核心代码