Foot only training

[Jaykob]

First of all, thanks for sharing your really valuable work with us!

I'm sorry if it has been asked before, but I'm a bit confused on how to tackle my problem.
I'm trying to train a foot only detector/estimator. It should be able to detect the foot keypoints in images that only contain the foot and a bit of the leg - no complete body.
I'm a bit confused on what to set in d_setLayers.py to only train the coco2017_foot dataset.
Is it possible at all or am I better off using CPM for that or sth. completely different?