| Safe Haskell | None |
|---|---|
| Language | Haskell2010 |
DataFrame.SVM.RFF
Description
Approximate RBF-kernel SVM via Random Fourier Features (Rahimi & Recht): map
each row through z(x) = √(2/D)·cos(W x + b) with W ~ N(0, 2γI) (seeded), then
fit a linear SVC in the random-feature space. predict compiles to a closed
Σ_r β_r·cos(…) expression of size O(D·d), independent of the row count.
Synopsis
- module DataFrame.Model
- data RFFConfig = RFFConfig {}
- defaultRFFConfig :: RFFConfig
- data RFFSVMModel a = RFFSVMModel {
- rffW :: !(Vector (Vector Double))
- rffB :: !(Vector Double)
- rffCoef :: !(Vector Double)
- rffIntercept :: !Double
- rffScale :: !Double
- rffNegClass :: !a
- rffPosClass :: !a
- rffFeatureNames :: !(Vector Text)
Documentation
module DataFrame.Model
Constructors
| RFFConfig | |
Instances
| Show RFFConfig Source # | |||||||||
| Eq RFFConfig Source # | |||||||||
| (Columnable a, Ord a) => Fit RFFConfig (Expr a) Source # | |||||||||
Defined in DataFrame.SVM.RFF Associated Types
| |||||||||
| type FrameReq RFFConfig (Expr a) Source # | |||||||||
Defined in DataFrame.SVM.RFF | |||||||||
| type ModelOf RFFConfig (Expr a) Source # | |||||||||
Defined in DataFrame.SVM.RFF | |||||||||
data RFFSVMModel a Source #
A fitted RFF SVM (binary). rffW / rffB are the random projection;
rffCoef / rffIntercept the linear SVC in feature space.
Constructors
| RFFSVMModel | |
Fields
| |
Instances
| Show a => Show (RFFSVMModel a) Source # | |||||
Defined in DataFrame.SVM.RFF Methods showsPrec :: Int -> RFFSVMModel a -> ShowS # show :: RFFSVMModel a -> String # showList :: [RFFSVMModel a] -> ShowS # | |||||
| Columnable a => Predict (RFFSVMModel a) Source # | |||||
Defined in DataFrame.SVM.RFF Associated Types
Methods predict :: RFFSVMModel a -> Prediction (RFFSVMModel a) Source # | |||||
| type Prediction (RFFSVMModel a) Source # | |||||
Defined in DataFrame.SVM.RFF | |||||