Skip to main content

Table 1 Comparison of Lasso, Random Forests, and MLP-SAE model

From: A deep auto-encoder model for gene expression prediction

Method

Hyperparameter

Hyperparameter value

MSE

Lasso

α

0.05

0.3516

  

0.1

0.3182

  

0.2

0.3002

  

0.3

0.2951

  

0.4

0.2930

  

0.5

0.2918

  

0.6

0.2914

  

0.7

0.2912

  

0.8

0.2912

Random forests

Number of estimators

10

0.3221

  

20

0.3127

  

30

0.3080

  

40

0.3001

  

50

0.2989

  

60

0.3003

  

70

0.2986

  

100

0.3003

  

150

0.2974

  

200

0.2967

MLP-SAE model

Learning rate

0.1

0.2890

  

0.01

0.2909

  

0.001

0.2895

  

0.0001

0.2908

  

0.00001

0.2918

  1. Each row represents the hyperparameter used and corresponding MSE for each hyperparameter setup of each model. Bold rows denote the hyperparameters and corresponding MSE for the optimal models of the three methods respectively