Struct PyARDRegression 

pub struct PyARDRegression { /* private fields */ }

Bayesian ARD regression.

Fit the weights of a regression model, using an ARD prior. The weights of the regression model are assumed to be drawn from an isotropic Gaussian distribution with precision lambda. The shrinkage is data-dependent, and the parameters of the prior are estimated from the data using empirical Bayes approach.

Parameters

max_iter : int, default=300 Maximum number of iterations.

tol : float, default=1e-3 Stop the algorithm if w has converged.

alpha_init : float, default=1.0 Initial value for alpha (per-feature precisions). If not provided, alpha_init is 1.0.

lambda_init : float, default=1.0 Initial value for lambda (precision of the noise). If not provided, lambda_init is 1.0.

threshold_alpha : float, default=1e10 Threshold for removing (pruning) weights with high precision from the computation: features with precision higher than this threshold are considered to have zero weight.

fit_intercept : bool, default=True Whether to calculate the intercept for this model. If set to False, no intercept will be used in calculations (i.e. data is expected to be centered).

compute_score : bool, default=False If True, compute the objective function at each step of the model.

copy_X : bool, default=True If True, X will be copied; else, it may be overwritten.

Attributes

coef_ : array-like of shape (n_features,) Coefficients of the regression model (mean of distribution)

alpha_ : array-like of shape (n_features,) estimated precision of the weights.

lambda_ : float estimated precision of the noise.

sigma_ : array-like of shape (n_features, n_features) estimated variance-covariance matrix of the weights

scores_ : array-like of shape (n_iter_+1,) if computed, value of the objective function (to be maximized) at each iteration of the optimization.

intercept_ : float Independent term in decision function. Set to 0.0 if fit_intercept = False.

n_features_in_ : int Number of features seen during fit.

Examples

from sklears_python import ARDRegression import numpy as np X = np.array([[1], [2], [3], [4], [5]]) y = np.array([1, 2, 3, 4, 5]) reg = ARDRegression() reg.fit(X, y) ARDRegression() reg.predict([[3]]) array([3.])

Notes

ARD performs feature selection by setting the weights of many features to zero, as they are deemed irrelevant. This is particularly useful when the number of features is much larger than the number of samples.

For polynomial regression, it is recommended to “center” the data by subtracting its mean before fitting the ARD model.

References

D. J. C. MacKay, Bayesian nonlinear modeling for the prediction competition, ASHRAE Transactions, 1994.

R. Salakhutdinov, Lecture notes on Statistical Machine Learning, http://www.cs.toronto.edu/~rsalakhu/sta4273/notes/Lecture2.pdf Their beta is our lambda_, and their alpha is our alpha_ ARD is a little different: only lambda_ is inferred; alpha_ is fixed by the user.

Trait Implementations

impl<'py> IntoPyObject<'py> for PyARDRegression

type Target = PyARDRegression

The Python output type

type Output = Bound<'py, <PyARDRegression as IntoPyObject<'py>>::Target>

The smart pointer type to use.

type Error = PyErr

The type returned in the event of a conversion error.

fn into_pyobject( self, py: Python<'py>, ) -> Result<<Self as IntoPyObject<'_>>::Output, <Self as IntoPyObject<'_>>::Error>

Performs the conversion.

impl PyClass for PyARDRegression

type Frozen = False

Whether the pyclass is frozen.

impl PyClassImpl for PyARDRegression

const IS_BASETYPE: bool = false

#[pyclass(subclass)]

const IS_SUBCLASS: bool = false

#[pyclass(extends=…)]

const IS_MAPPING: bool = false

#[pyclass(mapping)]

const IS_SEQUENCE: bool = false

#[pyclass(sequence)]

const IS_IMMUTABLE_TYPE: bool = false

#[pyclass(immutable_type)]

type BaseType = PyAny

Base class

type ThreadChecker = SendablePyClass<PyARDRegression>

This handles following two situations:

type PyClassMutability = <<PyAny as PyClassBaseType>::PyClassMutability as PyClassMutability>::MutableChild

Immutable or mutable

type Dict = PyClassDummySlot

Specify this class has #[pyclass(dict)] or not.

type WeakRef = PyClassDummySlot

Specify this class has #[pyclass(weakref)] or not.

type BaseNativeType = PyAny

The closest native ancestor. This is PyAny by default, and when you declare #[pyclass(extends=PyDict)], it’s PyDict.

fn items_iter() -> PyClassItemsIter

fn doc(py: Python<'_>) -> PyResult<&'static CStr>

Rendered class doc

fn lazy_type_object() -> &'static LazyTypeObject<Self>

fn dict_offset() -> Option<isize>

fn weaklist_offset() -> Option<isize>

impl PyClassNewTextSignature<PyARDRegression> for PyClassImplCollector<PyARDRegression>

fn new_text_signature(self) -> Option<&'static str>

impl<'a, 'py> PyFunctionArgument<'a, 'py, false> for &'a PyARDRegression

type Holder = Option<PyRef<'py, PyARDRegression>>

fn extract( obj: &'a Bound<'py, PyAny>, holder: &'a mut Self::Holder, ) -> PyResult<Self>

impl<'a, 'py> PyFunctionArgument<'a, 'py, false> for &'a mut PyARDRegression

type Holder = Option<PyRefMut<'py, PyARDRegression>>

fn extract( obj: &'a Bound<'py, PyAny>, holder: &'a mut Self::Holder, ) -> PyResult<Self>

impl PyMethods<PyARDRegression> for PyClassImplCollector<PyARDRegression>

fn py_methods(self) -> &'static PyClassItems

impl PyTypeInfo for PyARDRegression

const NAME: &'static str = "ARDRegression"

Class name.

const MODULE: Option<&'static str> = ::core::option::Option::None

Module name, if any.

fn type_object_raw(py: Python<'_>) -> *mut PyTypeObject

Returns the PyTypeObject instance for this type.

fn type_object(py: Python<'_>) -> Bound<'_, PyType>

Returns the safe abstraction over the type object.

fn is_type_of(object: &Bound<'_, PyAny>) -> bool

Checks if object is an instance of this type or a subclass of this type.

fn is_exact_type_of(object: &Bound<'_, PyAny>) -> bool

Checks if object is an instance of this type.

impl DerefToPyAny for PyARDRegression