Trained and validated on Sheep No. No. of patterns in the train dataset Tested on Sheep No. No. of patterns in the Test-set TP hits TN hits FP hits FN hits Sensitivity [%] Selectivity [%] Precision [%] Accuracy [%]
2,3,4,5,6,7 4567 1 443 169 270 4 0 100 97.7 100 99.1
1,3,4,5,6,7 4751 2 259 110 149 0 0 100 100 100 100
1,2,4,5,6,7 4731 3 279 82 196 1 0 100 98.8 100 99.6
1,2,3,5,6,7 3372 4 1638 823 812 1 2 99.8 99.9 99.8 99.8
1,2,3,4,6,7 4088 5 922 455 455 12 0 100 97.4 100 98.7
1,2,3,4,5,7 4466 6 544 231 312 0 1 99.6 100 99.7 99.8
1,2,3,4,5,6 4085 7 925 201 716 0 8 100 96.2 100 99.1
Overall performance of the 11 layers WF-CNN in the entire 6 hours 99.44±0.44
Table S5. Results of the WF-CNN classifier for post-HI spike transient identification in experimental data (entire 6 hours – 9 layers)