Table 3 Overall accuracy in terms of the ROC AUC score and individual (Twinned and Not twinned) accuracy for different BN models with optimized hyperparameters.

From: Application of machine learning to assess the influence of microstructure on twin nucleation in Mg alloys

Model

Dataset

ROC AUC

Twinned accuracy

Not twinned accuracy

BN1

AZ31

0.877

0.885

0.869

BN2

AZ31 MB (S_SF1)

0.878

0.893

0.862

BN3

AZ31 MB (Max_deltaBSF)

0.878

0.894

0.862

BN4

AZ31 [T_SF1 < 0.16]

0.639

0.417

0.862

BN5

Mg-1Al

0.626

0.629

0.622

BN6

AZ31 [T_SF1 < 0.16] MB

0.735

0.767

0.703

BN7

Mg-1Al MB

0.685

0.664

0.705

BN8

AZ31 [T_SF1 < 0.16] MB (Min_deltaGs)

0.686

0.667

0.704

BN9

Mg-1Al MB (Grain_size)

0.676

0.652

0.701

  1. The scores presented here are the mean over the 10 cross-validation tasks. The “individual” scores were obtained by calculating the ratio between correct predictions and total samples (i.e., a value of 0 would mean that all predictions were incorrect and a value of 1 that all predictions were correct). The models were trained with the BN implementation available in the pyAgrum Python package84.
Back to article page