Module

Learn.Unsupervised.OutlierDetection

Package
purescript-learn
Repository
carldata/purescript-learn

Outlier detection This model is based on the assumption that features are independed and normaly distributed

#Model Source

data Model

Model based on normal distribution for each feature

Instances

#train Source

train :: Matrix Number -> Model

Train model (estimate distribution parameters) for each feature.

#predict Source

predict :: Model -> Matrix Number -> Vector Number

Predict values based on learned model. This function will return probabilities of each feature vector

#predict1 Source

predict1 :: Model -> Vector Number -> Number

Predict value for a single sample