| Safe Haskell | None |
|---|---|
| Language | Haskell2010 |
DataFrame.KMeans
Description
k-means clustering (Lloyd's algorithm with k-means++ seeding and multiple
restarts). fit trains a KMeansModel (inspectable centres); predict is the
arg-min cluster assignment. Per-cluster distance features are available via
kmeansDistanceExprs / kmeansTransform.
Synopsis
- module DataFrame.Model
- data KMeansConfig = KMeansConfig {}
- defaultKMeansConfig :: KMeansConfig
- data KMeansModel = KMeansModel {}
- kmeansDistanceExprs :: KMeansModel -> [(Text, Expr Double)]
- kmeansTransform :: KMeansModel -> Transform
Documentation
module DataFrame.Model
data KMeansConfig Source #
Constructors
| KMeansConfig | |
Instances
| Show KMeansConfig Source # | |||||||||
Defined in DataFrame.KMeans Methods showsPrec :: Int -> KMeansConfig -> ShowS # show :: KMeansConfig -> String # showList :: [KMeansConfig] -> ShowS # | |||||||||
| Eq KMeansConfig Source # | |||||||||
Defined in DataFrame.KMeans | |||||||||
| Fit KMeansConfig [Expr Double] Source # | |||||||||
Defined in DataFrame.KMeans Associated Types
| |||||||||
| type FrameReq KMeansConfig [Expr Double] Source # | |||||||||
Defined in DataFrame.KMeans | |||||||||
| type ModelOf KMeansConfig [Expr Double] Source # | |||||||||
Defined in DataFrame.KMeans | |||||||||
data KMeansModel Source #
A fitted k-means model. kmCenters are sklearn's cluster_centers_.
Constructors
| KMeansModel | |
Instances
| Show KMeansModel Source # | |||||
Defined in DataFrame.KMeans Methods showsPrec :: Int -> KMeansModel -> ShowS # show :: KMeansModel -> String # showList :: [KMeansModel] -> ShowS # | |||||
| Predict KMeansModel Source # | |||||
Defined in DataFrame.KMeans Associated Types
Methods | |||||
| Eq KMeansModel Source # | |||||
Defined in DataFrame.KMeans | |||||
| type Prediction KMeansModel Source # | |||||
Defined in DataFrame.KMeans | |||||
kmeansDistanceExprs :: KMeansModel -> [(Text, Expr Double)] Source #
Per-cluster squared-distance expressions, named dist1, dist2, …
kmeansTransform :: KMeansModel -> Transform Source #
The per-cluster distance features as a composable fitted Transform.