Fig. 7: Decision surfaces of Bayesian Networks models. | npj Computational Materials

Fig. 7: Decision surfaces of Bayesian Networks models.

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

Fig. 7

Decision surfaces of different BN models trained on the (a, c) “reduced” AZ31 and (b, d) the Mg-1Al datasets. Models were trained using Min_deltaBSF and either (a, b) Grain_size or (c, d) Min_deltaGs. Blue and red colors identify the zones of the xy plane where the model will predict “Not twinned” or “Twinned”, respectively. For comparison, the boundary of the Mg-1Al models is drawn with a yellow line on top the decision surfaces of the AZ31 models.

Back to article page