Enum linfa::composing::platt_scaling::PlattError [−][src]
pub enum PlattError {
LineSearchNotConverged,
MaxIterReached,
MaxIterZero,
MinStepNegative(f32),
SigmaNegative(f32),
LinfaError(Error),
}
Expand description
Platt Newton’s method errors
Errors occur when setting invalid parameters or the optimization process fails.
Variants
MinStepNegative(f32)
Tuple Fields of MinStepNegative
0: f32
SigmaNegative(f32)
Tuple Fields of SigmaNegative
0: f32
LinfaError(Error)
Tuple Fields of LinfaError
0: Error
Trait Implementations
fn fit_with(
&self,
obj: O,
ds: &DatasetBase<Array2<F>, Array1<bool>>
) -> Result<Self::ObjectOut, PlattError>
fn fit_with(
&self,
obj: O,
ds: &DatasetBase<Array2<F>, Array1<bool>>
) -> Result<Self::ObjectOut, PlattError>
Calibrate another model with Platt scaling
This function takes another model and binary decision dataset and calibrates it to produce probability values. The returned model therefore implements the prediction trait for probability targets.
type ObjectIn = O
Auto Trait Implementations
impl RefUnwindSafe for PlattError
impl Send for PlattError
impl Sync for PlattError
impl Unpin for PlattError
impl UnwindSafe for PlattError
Blanket Implementations
Mutably borrows from an owned value. Read more