Journal of Cheminformatics

Table 3 Estimated linear and Random Forest model evaluation for aqueous solubility predictions- Morgan-Fingerprint method

From: Prediction of organic compound aqueous solubility using machine learning: a comparison study of descriptor-based and fingerprints-based models

		Training set			Test set
		R²	RMSE	MAE	R²	RMSE	MAE
2048 features	RF	0.96	0.35	0.23	0.81	0.80	0.55
2048 features	MLR	0.83	0.74	0.55	0.66	1.10	0.80
631 features	RF	0.96	0.35	0.23	0.81	0.84	0.57
631 features	MLR	0.77	0.88	0.66	0.75	0.94	0.69

Back to article page

ISSN: 1758-2946

Contact us

Submission enquiries: journalsubmissions@springernature.com