Evaluating Models with COCO Validation Set Through Python Wrapper API

[gndowns]

I'm wondering if there is any simple working example about how to generate the coco json files for the COCO validation set from within Python using the Python Wrapper. The only examples provided for generating coco JSON seem to use the compiled binaries, but I want to know how to do this from within a Python script. Documentation for the Python API seems virtually nonexistent, I am not even sure what parameters I would need to use in order to run the wrapper on the directory of coco validation images.

Please let me know if this is possible and if it is outlined somewhere. Also if there is readable documentation for the parameters for the Python wrapper anywhere.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[mahmoudmhashem]

I think this comment is useful "https://github.com/CMU-Perceptual-Computing-Lab/openpose/issues/738#issuecomment-849091807" check it out

for pythonAPI points25=datum.poseKeypoints[0] index=[0,1,2,3,4,5,6,7,9,10,11,12,13,14,15,16,17,18] #body25->coco18 format points18=points25[index]