I have 14 videos, and I’ve completed iterations 0,1, and 2 of training on them. Currently, I have the elbow/fingers/paw/nose/joystick labelled on the mouse reaching task. I am looking to add 2 new body parts (the mouse’s haunch and shoulder) to the training / videos I already have. I read that its possible to add more body parts after training, but I’m having trouble finding out how to do it. What steps should I be following to do this correctly?
This is described in the Nath et al. Nature Protocols manuscript (https://doi.org/10.1038/s41596-019-0176-0) where it states:
“Occasionally, the user might want to label additional body parts. In such a case, the user needs to append the new labels to the bodyparts list in the config.yaml file. Thereafter, the user can call the function label_frames and will be asked if he or she wants to display only the new labels or all labels before loading the frames. Saving the labels after all the images are labeled will append the new labels to the existing labeled dataset.”