PyNomaly icon indicating copy to clipboard operation
PyNomaly copied to clipboard

Published 20 hours ago • verified publisher icon vc1492a

Alter Fit Convention

Open vc1492a opened this issue 6 years ago • 2 comments

Provide parameters in LocalOutlierProbability() and provide data in fit() as opposed to providing data in LocalOutlierProbability() along with params. This is s.t. PyNomaly is more in line with scikit-learn and other popular libraries.

vc1492a avatar Nov 14 '17 16:11 vc1492a

More specifically and as an update, the following parameters could stay in the LocalOutlierProbability() method:

  • extent
  • n_neighbors

And any data-related parameters that may change from execution to execution (such in the case of inference) could instead be passed to the fit() method:

  • data
  • distance_matrix
  • neighbor_matrix
  • use_numba
  • progress_bar

vc1492a avatar Sep 30 '19 05:09 vc1492a

See NetworkX implementation.

vc1492a avatar Jul 09 '20 17:07 vc1492a