Closed rousseab closed 4 years ago
A first crack at producing a truncated confusion matrix and the distribution of angular errors. I also rapatriate the data splitting code so it is easy to have the same splits in the training and in the analysis.
I now see you approved @sblackburn86; sorry for asking again. I addressed your suggestions, so I'll merge now.
A first crack at producing a truncated confusion matrix and the distribution of angular errors. I also rapatriate the data splitting code so it is easy to have the same splits in the training and in the analysis.