Struct linfa_kernel::KernelParams

pub struct KernelParams<F, N = CommonNearestNeighbour> { /* fields omitted */ }

Defines the set of parameters needed to build a kernel

Implementations

impl<F, N> KernelParams<F, N>

pub fn method(self, method: KernelMethod<F>) -> Self

Setter for method, the inner product used by the kernel

pub fn kind(self, kind: KernelType) -> Self

Setter for kind, whether to construct a dense or sparse kernel

pub fn nn_algo(self, nn_algo: N) -> Self

Setter for nn_algo, nearest neighbour algorithm for calculating adjacency matrices

Trait Implementations

impl<F: Float, N: NearestNeighbour> Transformer<&'_ ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, KernelBase<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, CsMatBase<F, usize, Vec<usize, Global>, Vec<usize, Global>, Vec<F, Global>, usize>>> for KernelParams<F, N>

fn transform(&self, x: &Array2<F>) -> Kernel<F>

Builds a kernel from a view of the input data.

Parameters

• x: view of a matrix of records (#records, #features)

A kernel build from x according to the parameters on which this method is called

Panics

If the kernel type is Sparse and the number of neighbors specified is not between 1 and #records-1

impl<'a, F: Float, N: NearestNeighbour> Transformer<&'_ ArrayBase<ViewRepr<&'a F>, Dim<[usize; 2]>>, KernelBase<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, CsMatBase<F, usize, Vec<usize, Global>, Vec<usize, Global>, Vec<F, Global>, usize>>> for KernelParams<F, N>

fn transform(&self, x: &ArrayView2<'a, F>) -> Kernel<F>

Builds a kernel from a view of the input data.

Parameters

• x: view of a matrix of records (#records, #features)

A kernel build from x according to the parameters on which this method is called

Panics

If the kernel type is Sparse and the number of neighbors specified is not between 1 and #records-1

impl<'a, F: Float, L: 'a, T: AsTargets<Elem = L> + FromTargetArray<'a, L>, N: NearestNeighbour> Transformer<&'a DatasetBase<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, T>, DatasetBase<KernelBase<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, CsMatBase<F, usize, Vec<usize, Global>, Vec<usize, Global>, Vec<F, Global>, usize>>, <T as FromTargetArray<'a, L>>::View>> for KernelParams<F, N>

fn transform(
    &self,
    x: &'a DatasetBase<Array2<F>, T>
) -> DatasetBase<Kernel<F>, T::View>

Builds a new Dataset with the kernel as the records and the same targets as the input one.

Parameters

• x: A dataset with a matrix of records (#records, #features) and any targets

Returns

A new dataset with:

• records: a kernel build from x.records() according to the parameters on which this method is called
• targets: same as x.targets()

Panics

If the kernel type is Sparse and the number of neighbors specified is not between 1 and #records-1

impl<'a, 'b, F: Float, L: 'b, T: AsTargets<Elem = L> + FromTargetArray<'b, L>, N: NearestNeighbour> Transformer<&'b DatasetBase<ArrayBase<ViewRepr<&'a F>, Dim<[usize; 2]>>, T>, DatasetBase<KernelBase<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, CsMatBase<F, usize, Vec<usize, Global>, Vec<usize, Global>, Vec<F, Global>, usize>>, <T as FromTargetArray<'b, L>>::View>> for KernelParams<F, N>

fn transform(
    &self,
    x: &'b DatasetBase<ArrayView2<'a, F>, T>
) -> DatasetBase<Kernel<F>, T::View>

Builds a new Dataset with the kernel as the records and the same targets as the input one.

Parameters

• x: A dataset with a matrix of records (##records, ##features) and any targets

Returns

A new dataset with:

• records: a kernel build from x.records() according to the parameters on which this method is called
• targets: a slice of x.targets()

Panics

If the kernel type is Sparse and the number of neighbors specified is not between 1 and ##records-1

impl<'a, F: Float, N: NearestNeighbour> Transformer<ArrayBase<ViewRepr<&'a F>, Dim<[usize; 2]>>, KernelBase<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, CsMatBase<F, usize, Vec<usize, Global>, Vec<usize, Global>, Vec<F, Global>, usize>>> for KernelParams<F, N>

fn transform(&self, x: ArrayView2<'a, F>) -> Kernel<F>

Builds a kernel from a view of the input data.

Parameters

• x: view of a matrix of records (#records, #features)

A kernel build from x according to the parameters on which this method is called

Panics

If the kernel type is Sparse and the number of neighbors specified is not between 1 and #records-1

impl<'a, F: Float, T: AsTargets, N: NearestNeighbour> Transformer<DatasetBase<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, T>, DatasetBase<KernelBase<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, CsMatBase<F, usize, Vec<usize, Global>, Vec<usize, Global>, Vec<F, Global>, usize>>, T>> for KernelParams<F, N>

fn transform(&self, x: DatasetBase<Array2<F>, T>) -> DatasetBase<Kernel<F>, T>

Builds a new Dataset with the kernel as the records and the same targets as the input one.

It takes ownership of the original dataset.

Parameters

• x: A dataset with a matrix of records (#records, #features) and any targets

Returns

A new dataset with:

• records: a kernel build from x.records() according to the parameters on which this method is called
• targets: same as x.targets()

Panics

If the kernel type is Sparse and the number of neighbors specified is not between 1 and #records-1