pub trait Solve<A: Scalar> {
// Required methods
fn solve_inplace<'a>(
&self,
b: &'a mut ArrayRef<A, Ix1>,
) -> Result<&'a mut ArrayRef<A, Ix1>>;
fn solve_t_inplace<'a>(
&self,
b: &'a mut ArrayRef<A, Ix1>,
) -> Result<&'a mut ArrayRef<A, Ix1>>;
fn solve_h_inplace<'a>(
&self,
b: &'a mut ArrayRef<A, Ix1>,
) -> Result<&'a mut ArrayRef<A, Ix1>>;
// Provided methods
fn solve(&self, b: &ArrayRef<A, Ix1>) -> Result<Array1<A>> { ... }
fn solve_into<S: DataMut<Elem = A>>(
&self,
b: ArrayBase<S, Ix1>,
) -> Result<ArrayBase<S, Ix1>> { ... }
fn solve_t(&self, b: &ArrayRef<A, Ix1>) -> Result<Array1<A>> { ... }
fn solve_t_into<S: DataMut<Elem = A>>(
&self,
b: ArrayBase<S, Ix1>,
) -> Result<ArrayBase<S, Ix1>> { ... }
fn solve_h(&self, b: &ArrayRef<A, Ix1>) -> Result<Array1<A>> { ... }
fn solve_h_into<S: DataMut<Elem = A>>(
&self,
b: ArrayBase<S, Ix1>,
) -> Result<ArrayBase<S, Ix1>> { ... }
}Expand description
An interface for solving systems of linear equations.
There are three groups of methods:
solve*(normal) methods solveA * x = bforx.solve_t*(transpose) methods solveA^T * x = bforx.solve_h*(Hermitian conjugate) methods solveA^H * x = bforx.
Within each group, there are three methods that handle ownership differently:
*methods take a reference toband returnxas a new array.*_intomethods take ownership ofb, store the result in it, and return it.*_inplacemethods take a mutable reference toband store the result in that array.
If you plan to solve many equations with the same A matrix but different
b vectors, it’s faster to factor the A matrix once using the
Factorize trait, and then solve using the LUFactorized struct.
Required Methods§
Sourcefn solve_inplace<'a>(
&self,
b: &'a mut ArrayRef<A, Ix1>,
) -> Result<&'a mut ArrayRef<A, Ix1>>
fn solve_inplace<'a>( &self, b: &'a mut ArrayRef<A, Ix1>, ) -> Result<&'a mut ArrayRef<A, Ix1>>
Solves a system of linear equations A * x = b where A is self, b
is the argument, and x is the successful result.
§Panics
Panics if the length of b is not the equal to the number of columns
of A.
Provided Methods§
Sourcefn solve(&self, b: &ArrayRef<A, Ix1>) -> Result<Array1<A>>
fn solve(&self, b: &ArrayRef<A, Ix1>) -> Result<Array1<A>>
Solves a system of linear equations A * x = b where A is self, b
is the argument, and x is the successful result.
§Panics
Panics if the length of b is not the equal to the number of columns
of A.
Sourcefn solve_into<S: DataMut<Elem = A>>(
&self,
b: ArrayBase<S, Ix1>,
) -> Result<ArrayBase<S, Ix1>>
fn solve_into<S: DataMut<Elem = A>>( &self, b: ArrayBase<S, Ix1>, ) -> Result<ArrayBase<S, Ix1>>
Solves a system of linear equations A * x = b where A is self, b
is the argument, and x is the successful result.
§Panics
Panics if the length of b is not the equal to the number of columns
of A.
Sourcefn solve_t(&self, b: &ArrayRef<A, Ix1>) -> Result<Array1<A>>
fn solve_t(&self, b: &ArrayRef<A, Ix1>) -> Result<Array1<A>>
Solves a system of linear equations A^T * x = b where A is self, b
is the argument, and x is the successful result.
§Panics
Panics if the length of b is not the equal to the number of rows of
A.
Sourcefn solve_t_into<S: DataMut<Elem = A>>(
&self,
b: ArrayBase<S, Ix1>,
) -> Result<ArrayBase<S, Ix1>>
fn solve_t_into<S: DataMut<Elem = A>>( &self, b: ArrayBase<S, Ix1>, ) -> Result<ArrayBase<S, Ix1>>
Solves a system of linear equations A^T * x = b where A is self, b
is the argument, and x is the successful result.
§Panics
Panics if the length of b is not the equal to the number of rows of
A.
Dyn Compatibility§
This trait is not dyn compatible.
In older versions of Rust, dyn compatibility was called "object safety".