rlst 0.10.2

A Rust native linear algebra library.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157

//! Implementation of operator concepts for dense arrays.
//!
//! This module defines an [ArrayVectorSpace], a concrete implementation of [LinearSpace] based on
//! vectors represented through a 1d [Array](crate::dense::array::Array). A [ArrayVectorSpace] is also
//! an [InnerProductSpace] and a [NormedSpace](crate::NormedSpace). For the inner product the second vector
//! is taken as complex conjugate.

use std::{marker::PhantomData, ops::Neg};

use crate::{
    EvaluateArray, InnerProductSpace, LinearSpace, RlstScalar, dense::array::DynArray,
    operator::element::Element,
};

/// Finite-dimensional vector space backed by dynamic one-dimensional arrays.
///
/// Elements created by this space have the configured dimension. Arithmetic
/// assumes both operands came from a compatible space instance; mismatched
/// array lengths are rejected by the underlying array operations.
pub struct ArrayVectorSpace<Item> {
    dimension: usize,
    _marker: PhantomData<Item>,
}

impl<Item> ArrayVectorSpace<Item> {
    /// Create a vector space whose elements contain `dimension` scalars.
    pub fn new(dimension: usize) -> Self {
        Self {
            dimension,
            _marker: PhantomData,
        }
    }
}

impl<Item> LinearSpace for ArrayVectorSpace<Item>
where
    Item: Copy
        + Default
        + 'static
        + std::ops::Add<Output = Item>
        + std::ops::Sub<Output = Item>
        + std::ops::Neg<Output = Item>
        + std::ops::AddAssign<Item>
        + std::ops::SubAssign<Item>
        + std::ops::Mul<Item, Output = Item>
        + std::ops::MulAssign<Item>,
{
    type F = Item;

    type Impl = DynArray<Item, 1>;

    fn zero(&self) -> crate::operator::element::Element<'_, Self> {
        Element::new(self, DynArray::<Item, 1>::from_shape([self.dimension]))
    }

    fn add(
        &self,
        x: &crate::operator::element::Element<Self>,
        y: &crate::operator::element::Element<Self>,
    ) -> crate::operator::element::Element<'_, Self> {
        Element::new(self, (x.imp().r() + y.imp().r()).eval())
    }

    fn sub(
        &self,
        x: &crate::operator::element::Element<Self>,
        y: &crate::operator::element::Element<Self>,
    ) -> crate::operator::element::Element<'_, Self> {
        Element::new(self, (x.imp().r() - y.imp().r()).eval())
    }

    fn scalar_mul(
        &self,
        scalar: &Self::F,
        x: &crate::operator::element::Element<Self>,
    ) -> crate::operator::element::Element<'_, Self> {
        Element::new(
            self,
            DynArray::<_, 1>::new_from(&x.imp().r().scalar_mul(*scalar)),
        )
    }

    fn neg(
        &self,
        x: &crate::operator::element::Element<Self>,
    ) -> crate::operator::element::Element<'_, Self> {
        Element::new(self, x.imp().r().neg().eval())
    }

    fn sum_inplace(
        &self,
        x: &mut crate::operator::element::Element<Self>,
        y: &crate::operator::element::Element<Self>,
    ) {
        *x.imp_mut() += y.imp().r();
    }

    fn sub_inplace(
        &self,
        x: &mut crate::operator::element::Element<Self>,
        y: &crate::operator::element::Element<Self>,
    ) {
        *x.imp_mut() -= y.imp().r();
    }

    fn scale_inplace(&self, scalar: &Self::F, x: &mut crate::operator::element::Element<Self>) {
        *x.imp_mut() *= *scalar;
    }

    fn copy_from(
        &self,
        x: &crate::operator::element::Element<Self>,
    ) -> crate::operator::element::Element<'_, Self> {
        Element::new(self, x.imp().eval())
    }

    /// Compute `y += alpha * x` without forming the scaled element.
    fn axpy_inplace(
        &self,
        alpha: &Self::F,
        x: &crate::operator::element::Element<Self>,
        y: &mut crate::operator::element::Element<Self>,
    ) {
        *y.imp_mut() += x.imp().r().scalar_mul(*alpha);
    }
}

impl<Item> crate::IndexableSpace for ArrayVectorSpace<Item>
where
    Self: LinearSpace,
{
    fn dimension(&self) -> usize {
        self.dimension
    }
}

impl<Item> InnerProductSpace for ArrayVectorSpace<Item>
where
    Item: Copy
        + Default
        + RlstScalar
        + 'static
        + std::ops::Add<Output = Item>
        + std::ops::Sub<Output = Item>
        + std::ops::Neg<Output = Item>
        + std::ops::AddAssign<Item>
        + std::ops::SubAssign<Item>
        + std::ops::Mul<Item, Output = Item>
        + std::ops::MulAssign<Item>,
{
    fn inner_product(&self, x: &Element<Self>, other: &Element<Self>) -> Self::F {
        if x.imp().shape() == [0] && other.imp().shape() == [0] {
            return Item::zero();
        }
        x.imp().inner(other.imp()).unwrap()
    }
}