feaselbysfs4iris01.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 28 行

Construct 10 KNN models, each with up to 4 inputs selected from 4 candidates...

Selecting input 1:
Model 1/10:1 --> Recognition rate = 58.7%
Model 2/10:2 --> Recognition rate = 48.0%
Model 3/10:3 --> Recognition rate = 88.0%
Model 4/10:4 --> Recognition rate = 88.0%
Currently selected inputs: 3

Selecting input 2:
Model 5/10:3, 1 --> Recognition rate = 90.7%
Model 6/10:3, 2 --> Recognition rate = 90.7%
Model 7/10:3, 4 --> Recognition rate = 95.3%
Currently selected inputs: 3, 4

Selecting input 3:
Model 8/10:3, 4, 1 --> Recognition rate = 95.3%
Model 9/10:3, 4, 2 --> Recognition rate = 95.3%
Currently selected inputs: 3, 4, 1

Selecting input 4:
Model 10/10:3, 4, 1, 2 --> Recognition rate = 96.0%
Currently selected inputs: 3, 4, 1, 2

Overall maximal recognition rate = 96.0%.
Overall selected inputs: 3, 4, 1, 2