Skip to main content

Table 2 The performance of the MLPNNs to predict the SY of rapeseed using all measured traits

From: Application of machine learning algorithms and feature selection in rapeseedĀ (Brassica napus L.) breeding for seed yield

Algorithm

Activation function

Best hidden layers topology

Training

Testing

R2

RMSE

MAE

R2

RMSE

MAE

Multilayer Perceptron Neural Network (MLPNN)

Identity

5

0.840

0.260

0.200

0.832

0.292

0.233

Ā 

Logistic

5

0.760

0.319

0.244

0.816

0.306

0.208

Ā 

Tanh

4ā€“5ā€“5

0.857

0.246

0.188

0.827

0.295

0.234

Ā 

Relu

2-5-4-2Ā 

0.855

0.247

0.190

0.820

0.302

0.237

  1. R2 determination coefficient, RMSE root mean square error, MAE mean absolute error