JPlogP: an improved logP predictor trained using predicted data

Table 2 RMSE values for the grid-search comparing the maximum number of matches to use along with the minimum similarity threshold

Maximum examples	Minimum similarity threshold
Maximum examples	0.8	0.75	0.7	0.6	0.5
1	0.65	0.647	0.659	0.704	0.731
2	0.644	0.639	0.651	0.663	0.687
3	0.642	0.636	0.645	0.656	0.678
4	0.64	0.634	0.644	0.654	0.677
5	0.639	0.633	0.643	0.651	0.672
6	0.639	0.633	0.642	0.649	0.67

ISSN: 1758-2946