Hi Tom, The anomaly_score values are raw scores computed by the NuPIC detector, which are then post-processed to yield anomaly_likelihood values. When the likelihood is above the threshold (0.99999), the detector flags the data point as anomalous. For implementation details please see the info posted in the One Hot Gym Anomaly readme [1].
Best, Alex [1] https://github.com/numenta/nupic/tree/master/examples/opf/clients/hotgym/anomaly/one_gym#anomaly-likelihood Alexander Lavin Software Engineer Numenta
