solvr 0.2.0-beta.2

Advanced computing library for real-world problem solving - optimization, differential equations, interpolation, statistics, and more
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

//! CPU implementation of SLSQP.

use numr::error::Result;
use numr::runtime::cpu::{CpuClient, CpuRuntime};
use numr::tensor::Tensor;

use crate::optimize::constrained::impl_generic::slsqp_impl;
use crate::optimize::constrained::traits::{
    Bounds, ConstrainedOptions, ConstrainedResult, Constraint, SlsqpAlgorithms,
};
use crate::optimize::error::OptimizeResult;

impl SlsqpAlgorithms<CpuRuntime> for CpuClient {
    fn slsqp<F>(
        &self,
        f: F,
        x0: &Tensor<CpuRuntime>,
        constraints: &[Constraint<'_, CpuRuntime>],
        bounds: &Bounds<CpuRuntime>,
        options: &ConstrainedOptions,
    ) -> OptimizeResult<ConstrainedResult<CpuRuntime>>
    where
        F: Fn(&Tensor<CpuRuntime>) -> Result<f64>,
    {
        slsqp_impl(self, f, x0, constraints, bounds, options)
    }
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::optimize::constrained::traits::ConstraintType;
    use numr::runtime::cpu::CpuDevice;

    fn setup() -> (CpuDevice, CpuClient) {
        let device = CpuDevice::new();
        let client = CpuClient::new(device.clone());
        (device, client)
    }

    #[test]
    fn test_slsqp_unconstrained() {
        let (device, client) = setup();
        let x0 = Tensor::<CpuRuntime>::from_slice(&[2.0, 3.0], &[2], &device);

        let result = client
            .slsqp(
                |x| {
                    let d: Vec<f64> = x.to_vec();
                    Ok(d.iter().map(|xi| xi * xi).sum())
                },
                &x0,
                &[],
                &Bounds::default(),
                &ConstrainedOptions::default(),
            )
            .unwrap();

        assert!(result.converged);
        assert!(result.fun < 1e-4);
    }

    #[test]
    fn test_slsqp_with_bounds() {
        let (device, client) = setup();
        let x0 = Tensor::<CpuRuntime>::from_slice(&[2.0, 3.0], &[2], &device);
        let lower = Tensor::<CpuRuntime>::from_slice(&[1.0, 1.0], &[2], &device);
        let upper = Tensor::<CpuRuntime>::from_slice(&[10.0, 10.0], &[2], &device);

        let bounds = Bounds {
            lower: Some(lower),
            upper: Some(upper),
        };

        let result = client
            .slsqp(
                |x| {
                    let d: Vec<f64> = x.to_vec();
                    Ok(d.iter().map(|xi| xi * xi).sum())
                },
                &x0,
                &[],
                &bounds,
                &ConstrainedOptions::default(),
            )
            .unwrap();

        // Should find a solution within bounds
        let sol: Vec<f64> = result.x.to_vec();
        assert!(sol[0] >= 1.0 - 1e-6);
        assert!(sol[1] >= 1.0 - 1e-6);
        assert!(sol[0] <= 10.0 + 1e-6);
        assert!(sol[1] <= 10.0 + 1e-6);
    }

    #[test]
    fn test_slsqp_equality_constraint() {
        let (device, client) = setup();
        // Minimize x^2 + y^2 subject to x + y = 1
        let x0 = Tensor::<CpuRuntime>::from_slice(&[0.5, 0.5], &[2], &device);

        let constraint = Constraint {
            kind: ConstraintType::Equality,
            fun: &|x: &Tensor<CpuRuntime>| {
                let d: Vec<f64> = x.to_vec();
                let val = d[0] + d[1] - 1.0;
                Ok(Tensor::<CpuRuntime>::from_slice(&[val], &[1], x.device()))
            },
            jac: None,
        };

        let result = client
            .slsqp(
                |x| {
                    let d: Vec<f64> = x.to_vec();
                    Ok(d.iter().map(|xi| xi * xi).sum())
                },
                &x0,
                &[constraint],
                &Bounds::default(),
                &ConstrainedOptions {
                    max_iter: 200,
                    ..Default::default()
                },
            )
            .unwrap();

        // Minimum at (0.5, 0.5) with f = 0.5
        let sol: Vec<f64> = result.x.to_vec();
        assert!(
            (sol[0] - 0.5).abs() < 0.1,
            "x[0] = {} expected ~0.5",
            sol[0]
        );
        assert!(
            (sol[1] - 0.5).abs() < 0.1,
            "x[1] = {} expected ~0.5",
            sol[1]
        );
    }
}