A Novel Molecular Representation Learning for Molecular Property Prediction with a Multiple SMILES-Based Augmentation

<div>Scatter diagram of FreeSolv dataset for four training folds. The horizontal axis is the ground truth value, and the vertical axis is the predicted value by our model. The solid lines indicate the trend lines, and the dashed lines indicate the identity lines. The blue and red points represent the predicted values in the validation set and test set, respectively.</div>

Computational Intelligence and Neuroscience

fig9

Figure 9

Figure 9: A Novel Molecular Representation Learning for Molecular Property Prediction with a Multiple SMILES-Based Augmentation 