Struct argmin::core::test_utils::TestSparseProblem

source · [−]

pub struct TestSparseProblem {}

Expand description

A struct representing the following sparse problem.

Example 1: x = [1, 1, 0, 0], y = 1 Example 2: x = [0, 0, 1, 1], y = -1 Example 3: x = [1, 0, 0, 0], y = 1 Example 4: x = [0, 0, 1, 0], y = -1

cost = Σ (w^T x - y)^2

Implements CostFunction and Gradient.

Implementations

source

impl TestSparseProblem

source

pub fn new() -> Self

Create an instance of TestSparseProblem.

Example

use argmin::core::test_utils::TestSparseProblem;

let problem = TestSparseProblem::new();

Trait Implementations

source

impl Clone for TestSparseProblem

source

fn clone(&self) -> TestSparseProblem

Returns a copy of the value. Read more

1.0.0 · source

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source. Read more

source

impl CostFunction for TestSparseProblem

source

fn cost(&self, param: &Self::Param) -> Result<Self::Output, Error>

Returns a sum of squared errors.

Example

use argmin::core::test_utils::TestSparseProblem;
use argmin::core::CostFunction;

let problem = TestSparseProblem::new();

let param = vec![1.0, 2.0, 3.0, 4.0];

let res = problem.cost(&param)?;

type Output = f64

Tyope of the return value of the cost function

source

fn bulk_cost<'a, P>(&self, params: &'a [P]) -> Result<Vec<Self::Output>, Error> where
 P: Borrow<Self::Param> + SyncAlias,
 Self::Output: SendAlias,
 Self: SyncAlias,

Compute cost in bulk. If the rayon feature is enabled, multiple calls to cost will be run in parallel using rayon, otherwise they will execute sequentially. If the rayon feature is enabled, parallelization can still be turned off by overwriting parallelize to return false. This can be useful in cases where it is preferable to parallelize only certain parts. Note that even if parallelize is set to false, the parameter vectors and the problem are still required to be Send and Sync. Those bounds are linked to the rayon feature. This method can be overwritten.

source

fn parallelize(&self) -> bool

Indicates whether to parallelize calls to cost when using bulk_cost. By default returns true, but can be set manually to false if needed. This allows users to turn off parallelization for certain traits implemented on their problem. Note that parallelization requires the rayon feature to be enabled, otherwise calls to cost will be executed sequentially independent of how parallelize is set.

source

impl Debug for TestSparseProblem

source

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter. Read more

source

impl Default for TestSparseProblem

source

fn default() -> TestSparseProblem

Returns the “default value” for a type. Read more

source

impl<'de> Deserialize<'de> for TestSparseProblem

source

fn deserialize<D>(deserializer: D) -> Result<Self, D::Error> where
__D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer. Read more

source

impl Gradient for TestSparseProblem

source

fn gradient(&self, param: &Self::Param) -> Result<Self::Gradient, Error>

Returns a gradient of the cost function.

Example

use argmin::core::test_utils::TestSparseProblem;
use argmin::core::Gradient;

let problem = TestSparseProblem::new();

let param = vec![1.0, 2.0, 3.0, 4.0];

let res = problem.gradient(&param)?;

type Param = Vec<f64, Global>

Type of the parameter vector

type Gradient = Vec<f64, Global>

Type of the gradient

source

fn bulk_gradient<'a, P>(
 &self,
 params: &'a [P]
) -> Result<Vec<Self::Gradient>, Error> where
 P: Borrow<Self::Param> + SyncAlias,
 Self::Gradient: SendAlias,
 Self: SyncAlias,

Compute gradient in bulk. If the rayon feature is enabled, multiple calls to gradient will be run in parallel using rayon, otherwise they will execute sequentially. If the rayon feature is enabled, parallelization can still be turned off by overwriting parallelize to return false. This can be useful in cases where it is preferable to parallelize only certain parts. Note that even if parallelize is set to false, the parameter vectors and the problem are still required to be Send and Sync. Those bounds are linked to the rayon feature. This method can be overwritten.

source

fn parallelize(&self) -> bool

Indicates whether to parallelize calls to gradient when using bulk_gradient. By default returns true, but can be set manually to false if needed. This allows users to turn off parallelization for certain traits implemented on their problem. Note that parallelization requires the rayon feature to be enabled, otherwise calls to gradient will be executed sequentially independent of how parallelize is set.

source