Error while setting metric to Euclidean

[sidkoch]

When i try this :

feature_select = NCFS(metric='euclidean') feature_select.fit(X, y)

I get an error (internally when it calls accelerated.fit())

SystemError: CPUDispatcher(<function fit at 0x7fcff95909d0>) returned a result with an error set

[sidkoch]

Is there a functionality to externally set parallel to False ?

[dakota-hawkins]

Weird! I'll try and check it out when I get the chance. It works with metric="manhattan"?

You can change the number of threads provided to it, which should essentially turn it off.

Info on limiting threads: https://numba.pydata.org/numba-doc/latest/user/threading-layer.html#setting-the-number-of-threads

In general, I would recommend not turning parallel off, as I can't guarantee how long it would take to converge.