HTM.java needs a KNNClassifier

[cogmission]

The KNNClassifier is required in order to have a full set of classifiers. In order to complete this, the swig wrapper class mapping of method implementations need to be identified. Namely these:

NearestNeighbor32: <-- I need to be able to view the implementation of these methods which I haven't been able to find in the NearestNeighbor.hpp file or any of their includes. I think that they are given aliases by swig generation or something? Anyway, if someone can point me to these I can begin work.

def rowDist(*args, **kwargs):
    """rowDist(self, row, xIn) -> Real"""
    return _math._NearestNeighbor32_rowDist(*args, **kwargs)

def vecLpDist(*args, **kwargs):
    """vecLpDist(self, p, xIn, take_root=True) -> PyObject"""
    return _math._NearestNeighbor32_vecLpDist(*args, **kwargs)

def LpNearest(*args, **kwargs):
    """LpNearest(self, p, row, k=1, take_root=True) -> PyObject"""
    return _math._NearestNeighbor32_LpNearest(*args, **kwargs)

def closestLp_w(*args, **kwargs):
    """closestLp_w(self, p, row) -> PyObject"""
    return _math._NearestNeighbor32_closestLp_w(*args, **kwargs)

def closestDot(*args, **kwargs):
    """closestDot(self, row) -> PyObject"""
    return _math._NearestNeighbor32_closestDot(*args, **kwargs)

def projLpNearest(*args, **kwargs):
    """projLpNearest(self, p, py_x, k=1, take_root=False) -> PyObject"""
    return _math._NearestNeighbor32_projLpNearest(*args, **kwargs)

def projRbf(*args, **kwargs):
    """projRbf(self, p, k, py_x) -> PyObject"""
    return _math._NearestNeighbor32_projRbf(*args, **kwargs)

[cogmission]

@scottpurdy @rcrowder @breznak @chetan51 @subutai @rhyolight

Can some C++ savvy individual help with the above identification of source?

[rcrowder]

Those look like part of the sparse matrix bindings -
https://github.com/numenta/nupic/blob/master/nupic/bindings/sparse_matrix.i#L2292

Line 2355 for rowDist, etc.