Skip to content



Lightweight Online Detector of Anomalies uses a collection of sparse random projection vectors to provide scalar inputs to an ensemble of unique one-dimensional equi-width histograms. Each histogram then estimates the probability density of the unknown sample using a limited feature set. The final predictions are derived from the averaged densities over the entire ensemble.

Interfaces: Estimator, Learner, Online, Scoring, Persistable

Data Type Compatibility: Continuous


# Name Default Type Description
1 estimators 100 int The number of projection/histogram pairs in the ensemble.
2 bins null int The number of equi-width bins for each histogram. If null then will estimate bin count.
3 contamination 0.1 float The proportion of outliers that are assumed to be present in the training set.


use Rubix\ML\AnomalyDetectors\Loda;

$estimator = new Loda(250, 8, 0.01);

Additional Methods#

This estimator does not have any additional methods.


  1. T. Pevný. (2015). Loda: Lightweight on-line detector of anomalies. 

  2. L. Birg´e et al. (2005). How Many Bins Should Be Put In A Regular Histogram.