Table 6 The table shows the test MAE of the SC model, proposed TL model and % error change for each of the target materials properties for prediction task of ‘Experimental Data’.

From: Structure-aware graph neural network based deep transfer learning framework for enhanced predictive analytics on diverse materials datasets

Dataset

Property

Data Size

Base

MAE of SC Model

MAE of Proposed TL Model

% Error Change

EFE

Deltae (eVatom−1)

605

0.2624

0.0894

0.0782

-12.53

KEFE

Deltae (eVatom−1)

1557

0.9375

0.1109

0.0608

-45.18

KEB

Bg (eV)

2432

1.3142

0.4085

0.3989

-2.35

  1. The lowest MAE values in each row are highlighted in bold.
Back to article page