Table 2. Comparison of across-animal model performance on additional dataset.

Table lists reported mean accuracy of different models evaluated on the same publicly accessible dataset from Chaudhary et al., 2021. We note in the text an assumption needed to compare these methods. $N$ indicates the number of template-test pairs used to calculate accuracy. (CRF method uses an atlas as the template, whereas we randomly take one of the nine individuals and designate that as the template). CPD and fDNC performance on this dataset are also shown in Figure 4F.

Method	Accuracy	$N$	Reported in
CPD	59%	8	This work
CRF (open atlas)	$\approx 40$%	9	Chaudhary et al., 2021
CRF (data driven atlas)	74%	9	Chaudhary et al., 2021
fDNC	78%	8	This work