Fig. 2: Distribution of predictions across all cross-validation folds and random seeds per compound in the blind testing set, for neural networks trained on the original feature set FS60 and dataset DS60. | npj Materials Degradation

Fig. 2: Distribution of predictions across all cross-validation folds and random seeds per compound in the blind testing set, for neural networks trained on the original feature set FS60 and dataset DS60.

From: Searching the chemical space for effective magnesium dissolution modulators: a deep learning approach using sparse features

Fig. 2

Boxes are coloured according to the compound’s mean predicted IE values in %. Compounds are sorted by descending mean experimental IE values, which are depicted as coloured diamonds.

Back to article page