featureselectexhaustive03.txt

来自「这是台湾张智星的模式识别的课件的源码」· 文本 代码 · 共 21 行

Construct 15 KNN models, each with up to 4 inputs selected from 4 candidates...
modelIndex 1/15: sepal length --> Recognition rate = 58.666667%
modelIndex 2/15: sepal width --> Recognition rate = 48.000000%
modelIndex 3/15: petal length --> Recognition rate = 88.000000%
modelIndex 4/15: petal width --> Recognition rate = 88.000000%
modelIndex 5/15: sepal length, sepal width --> Recognition rate = 71.333333%
modelIndex 6/15: sepal length, petal length --> Recognition rate = 91.333333%
modelIndex 7/15: sepal length, petal width --> Recognition rate = 93.333333%
modelIndex 8/15: sepal width, petal length --> Recognition rate = 89.333333%
modelIndex 9/15: sepal width, petal width --> Recognition rate = 92.000000%
modelIndex 10/15: petal length, petal width --> Recognition rate = 95.333333%
modelIndex 11/15: sepal length, sepal width, petal length --> Recognition rate = 88.666667%
modelIndex 12/15: sepal length, sepal width, petal width --> Recognition rate = 92.666667%
modelIndex 13/15: sepal length, petal length, petal width --> Recognition rate = 94.666667%
modelIndex 14/15: sepal width, petal length, petal width --> Recognition rate = 93.333333%
modelIndex 15/15: sepal length, sepal width, petal length, petal width --> Recognition rate = 94.666667%

Overall max recognition rate = 95.3%.
Overall selected inputs: petal length, petal width