| Safe Haskell | None |
|---|---|
| Language | Haskell2010 |
DataFrame.DBSCAN
Description
Density-based clustering (DBSCAN). Brute-force O(n²) region queries, no
spatial index — suitable for the in-memory scales this library targets. DBSCAN
is transductive: it has a Fit instance but deliberately no Predict instance
(there is no honest single prediction expression). dbscanSurrogateExpr fits an
interpretable decision-tree surrogate on the cluster labels instead.
Synopsis
- module DataFrame.Model
- data DBSCANConfig = DBSCANConfig {
- dbEps :: !Double
- dbMinSamples :: !Int
- defaultDBSCANConfig :: DBSCANConfig
- data DBSCANModel = DBSCANModel {
- dbLabels :: !(Vector Int)
- dbCoreSampleIndices :: !(Vector Int)
- dbNClusters :: !Int
- dbscanSurrogateExpr :: TreeConfig -> [Expr Double] -> DBSCANModel -> DataFrame -> Expr Int
Documentation
module DataFrame.Model
data DBSCANConfig Source #
Constructors
| DBSCANConfig | |
Fields
| |
Instances
| Show DBSCANConfig Source # | |||||||||
Defined in DataFrame.DBSCAN Methods showsPrec :: Int -> DBSCANConfig -> ShowS # show :: DBSCANConfig -> String # showList :: [DBSCANConfig] -> ShowS # | |||||||||
| Eq DBSCANConfig Source # | |||||||||
Defined in DataFrame.DBSCAN | |||||||||
| Fit DBSCANConfig [Expr Double] Source # | |||||||||
Defined in DataFrame.DBSCAN Associated Types
| |||||||||
| type FrameReq DBSCANConfig [Expr Double] Source # | |||||||||
Defined in DataFrame.DBSCAN | |||||||||
| type ModelOf DBSCANConfig [Expr Double] Source # | |||||||||
Defined in DataFrame.DBSCAN | |||||||||
data DBSCANModel Source #
A fitted DBSCAN labelling. dbLabels uses -1 for noise (sklearn's
labels_); dbCoreSampleIndices are the core points.
Constructors
| DBSCANModel | |
Fields
| |
Instances
| Show DBSCANModel Source # | |
Defined in DataFrame.DBSCAN Methods showsPrec :: Int -> DBSCANModel -> ShowS # show :: DBSCANModel -> String # showList :: [DBSCANModel] -> ShowS # | |
| Eq DBSCANModel Source # | |
Defined in DataFrame.DBSCAN | |
dbscanSurrogateExpr :: TreeConfig -> [Expr Double] -> DBSCANModel -> DataFrame -> Expr Int Source #
Fit a decision-tree surrogate on the DBSCAN labels so new rows can be
assigned an (approximate) cluster. Noise (-1) is its own class.