smile.anomaly
Anomaly and outlier detection.
Attributes
Members list
Value members
Concrete methods
Fits an Isolation Forest model for anomaly detection.
Fits an Isolation Forest model for anomaly detection.
Value parameters
- extensionLevel
-
extension level for hyperplanes (0 for standard axis-aligned).
- maxDepth
-
the maximum depth of an isolation tree (0 for log2(subsample)).
- subsample
-
the sampling rate for each tree.
- trees
-
the number of trees.
- x
-
training samples.
Attributes
- Returns
-
the fitted model.
Fits a Local Outlier Factor (LOF) model using a KD-tree for spatial neighborhood search.
Fits a Local Outlier Factor (LOF) model using a KD-tree for spatial neighborhood search.
Value parameters
- k
-
the number of nearest neighbors.
- x
-
training observations.
Attributes
- Returns
-
the fitted model.
Fits a one-class support vector machine (One-Class SVM).
Fits a one-class support vector machine (One-Class SVM).
Type parameters
- T
-
input observation type.
Value parameters
- kernel
-
the Mercer kernel function.
- nu
-
upper bound on the fraction of outliers.
- tol
-
tolerance of convergence test.
- x
-
training samples.
Attributes
- Returns
-
the fitted model.