jaime 2.3.1

j.a.i.m.e. is an ergonomic all purpose gradient descent engine
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

extern crate test;

mod polinomial_test {
    use std::ops::{Add, Mul};

    use crate::trainer::{
        adam_trainer::AdamTrainer,
        asymptotic_gradient_descent_trainer::AsymptoticGradientDescentTrainer,
        default_param_translator, genetic_trainer::GeneticTrainer, DataPoint, Trainer,
    };

    fn base_func(x: f32) -> f32 {
        1. * (x * x * x * x * x) - 4. * (x * x * x * x) - 10. * (x * x * x)
            + 40. * (x * x)
            + 9. * x
            + -11.
    }

    #[test]
    fn genetic_polinomial_test() {
        polinomial_test(
            &mut GeneticTrainer::<_, _, _, 100, 5, _, _, _>::new(
                polinomial::<6, _>,
                default_param_translator,
                (),
                0.01,
                10,
            ),
            0.1,
        );
    }

    #[test]
    fn gradient_descent_polinomial_test() {
        polinomial_test(
            &mut AsymptoticGradientDescentTrainer::new_dense(
                polinomial::<6, _>,
                polinomial::<6, _>,
                default_param_translator,
                (),
            ),
            1.,
        );
    }

    #[test]
    fn adam_polinomial_test() {
        polinomial_test(
            &mut AdamTrainer::new_dense(
                polinomial::<6, _>,
                polinomial::<6, _>,
                default_param_translator,
                (),
                0.0001,
                10,
            ),
            0.001,
        );
    }

    fn polinomial_test<T: Trainer<6, 1, 1>>(trainer: &mut T, learning_rate: f32) {
        let mut complexity = 0.;

        while complexity < 1. {
            let dataset = dataset_service(complexity);

            while !trainer.found_local_minima() {
                trainer.train_step::<true, false, _, _>(
                    &dataset,
                    &dataset,
                    dataset.len(),
                    dataset.len(),
                    learning_rate,
                )
            }

            complexity += 0.1;
        }

        let cost = trainer.get_last_cost().unwrap();
        assert!(cost < 0.01);

        let params = trainer.get_model_params();
        let diff = params
            .iter()
            .zip([-11., 9., 40., -10., -4., 1.].iter())
            .fold(0., |acc, (a, b)| acc + a - b);
        assert!(diff < 0.01);
    }

    fn dataset_service<const P: usize>(complexity: f32) -> Vec<DataPoint<P, 1, 1>> {
        let abs_max = (complexity + 1.) * 100.;

        (-(abs_max) as isize..(abs_max) as isize)
            .map(|x| x as f32 / 100.)
            // .map(|x| x as f32 / 10.)
            .map(|x| DataPoint {
                input: [x],
                output: [base_func(x)],
            })
            .collect::<Vec<_>>()
    }

    pub fn polinomial<
        const G: usize,
        N: Clone
            + From<f32>
            + PartialOrd<f32>
            + PartialOrd<N>
            + Add<N, Output = N>
            + Mul<N, Output = N>
            + Mul<f32, Output = N>,
    >(
        params: &[N; G],
        input: &[f32; 1],
        _: &(),
    ) -> [N; 1] {
        let mut ret = N::from(0.);
        let mut x_to_the_nth = N::from(1.);

        for n in 0..G {
            ret = ret + (x_to_the_nth.clone() * params[n].clone());

            x_to_the_nth = x_to_the_nth * input[0];
        }

        [ret]
    }
}