I use opencv3 in python.
I confirmation this page ↓
http://docs.opencv.org/3.0-beta/doc/py_tutorials/py_ml/py_kmeans/py_kmeans_opencv/py_kmeans_opencv.html
In this explain , cv2.kmeans() function required 5 params 'samples', 'nclusters(K)', 'criteria', 'attempts', 'flags'.
but there examples required 6 params. 'None' is in third param.
I found cv2.kmeans() seems required 'bestLabels" parameter.
In many cases , it is None.
what is the 'bestLabels' ?
is it case of non-"None"?