Table 2.

The precision@k, recall@k, and F1 score@K of structure annotation from different versions of DeepSAT compared with SMART 2.0.

Model	Precision@k			Recall@k			F1 score@K
	k = 1	k = 5	k = 10	k = 1	k = 5	k = 10	k = 1	k = 5	k = 10
DeepSAT (multiplicty edited HSQC)	0.7351	0.6283	0.5358	0.1289	0.3917	0.5349	0.2193	0.4826	0.5353
DeepSAT (normal HSQC)	0.7037	0.6056	0.5232	0.1153	0.3585	0.5036	0.1981	0.4504	0.5132
SMART 2.0	0.4032	0.2800	0.2225	0.0596	0.1565	0.2131	0.1038	0.2008	0.2177

Best values in each column are bolded