Table 3 Hyperparameter of ML algorithm obtained as a result of TPE.

From: An efficient bearing fault detection strategy based on a hybrid machine learning technique

 

Hyperparameters

CNN models for feature extraction

Densenet201

Vgg16

Vgg19

MobileNetv2

Inceptionv3

ResNet50

Inceptionresnetv2

SVM

c

0.080756609525

0.333255597347

50.49285933919

0.183770163831

0.056487139653

0.069750972994

47.96146722467

Gamma

0.153368758782

0.201320138608

0.010661107491

0.136295952469

0.016368895980

0.053690736485

0.939256254370

Kernel

Linear

Linear

LINEAR

LINEAR

LINEAR

LINEAR

Linear

KNN

n_neighbors

3

19

8

7

11

5

10

p

1

1

1

2

1

1

2

DT

max_depth

17

14

12

11

14

13

18

min_samples_split

3

18

4

8

14

2

10

min_samples_leaf’

8

4

7

4

14

16

18

RF

n_estimators

170

170

190

180

150

100

140

max_depth

19

17

12

14

9

20

14

min_samples_split

14

14

11

11

2

5

18

min_samples_leaf

9

8

3

6

8

3

7

Back to article page