. Author manuscript; available in PMC: 2022 Nov 1.

Published in final edited form as: Mol Ecol Resour. 2021 Mar 8;21(8):2676–2688. doi: 10.1111/1755-0998.13355

Table 2:

Precision and recall of HyDe-CNN trained on images of minimum d_XY for model selection on independent test data. Divergence scaling is given in coalescent units (CU).

Divergence Scaling	Model	Precision	Recall
0.5	no_hyb	0.841	0.924
	hyb_sp	0.855	0.882
	admix	0.777	0.640
	admix_mig	0.846	0.885
1.0	no_hyb	0.928	0.965
	hyb_sp	0.886	0.942
	admix	0.919	0.774
	admix_mig	0.942	0.995
2.0	no_hyb	0.945	0.980
	hyb_sp	0.979	0.880
	admix	0.874	0.916
	admix_mig	0.983	1.000