puffpastry 0.1.0

basic rust neural network framework
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

use std::fmt::Debug;
use std::ops::{Add, Div, Mul, Sub};

pub trait ValidNumber<T>:
    Copy
    + Add<T, Output = T>
    + Mul<T, Output = T>
    + Sub<T, Output = T>
    + Div<T, Output = T>
    + From<f64>
    + Into<f64>
    + Debug
    + PartialOrd
{
}

impl<T> ValidNumber<T> for T where
    T: Copy
        + Add<T, Output = T>
        + Mul<T, Output = T>
        + Sub<T, Output = T>
        + Div<T, Output = T>
        + From<f64>
        + Into<f64>
        + Debug
        + PartialOrd
{
}

pub trait AddVec<T> {
    fn add_vec(&self, other: &Vec<T>) -> Vec<T>;
}

impl<T: ValidNumber<T>> AddVec<T> for Vec<T> {
    fn add_vec(&self, other: &Vec<T>) -> Vec<T> {
        self.iter()
            .zip(other.iter())
            .map(|(x, y)| *x + *y)
            .collect()
    }
}

pub trait ElementwiseMatrixMultiply<T> {
    fn elementwise_multiply(&self, other: &Vec<Vec<T>>) -> Vec<Vec<T>>;
}

impl<T: ValidNumber<T>> ElementwiseMatrixMultiply<T> for Vec<Vec<T>> {
    fn elementwise_multiply(&self, other: &Vec<Vec<T>>) -> Vec<Vec<T>> {
        let mut out = vec![vec![T::from(0.0); self[0].len()]; self.len()];

        for row in 0..out.len() {
            for col in 0..out[row].len() {
                out[row][col] = self[row][col] * other[row][col];
            }
        }

        out
    }
}

pub trait DotProduct<T> {
    fn dot_product(&self, right: &Vec<T>) -> T;
}

impl<T: ValidNumber<T>> DotProduct<T> for Vec<T> {
    fn dot_product(&self, right: &Vec<T>) -> T {
        self.iter()
            .zip(right.iter())
            .fold(T::from(0.0), |acc, (x, y)| acc + (*x) * (*y))
    }
}

pub trait Transpose<T> {
    fn transposed(&self) -> Vec<Vec<T>>;
}

impl<T: ValidNumber<T>> Transpose<T> for Vec<Vec<T>> {
    fn transposed(&self) -> Vec<Vec<T>> {
        let mut out = vec![vec![self[0][0]; self.len()]; self[0].len()];

        for row in 0..self.len() {
            for col in 0..self[row].len() {
                out[col][row] = self[row][col];
            }
        }

        out
    }
}

pub trait MatrixMultiply<T> {
    fn matrix_multiply(&self, other: &Vec<Vec<T>>) -> Vec<Vec<T>>;
}

impl<T: ValidNumber<T>> MatrixMultiply<T> for Vec<Vec<T>> {
    fn matrix_multiply(&self, other: &Vec<Vec<T>>) -> Vec<Vec<T>> {
        let mut out: Vec<Vec<T>> = vec![vec![other[0][0]; other[0].len()]; self.len()];
        let other = other.transposed();

        for row in 0..out.len() {
            for col in 0..out[0].len() {
                out[row][col] = self[row].dot_product(&other[col]);
            }
        }

        out
    }
}

mod test {
    use super::*;

    #[test]
    fn mat_mul_test_1() {
        let mat1 = vec![vec![1.0, 0.0], vec![0.0, 1.0]];
        let mat2 = vec![vec![1.0], vec![2.0]];
        let res = mat1.matrix_multiply(&mat2);

        assert_eq!(res, mat2);
    }

    #[test]
    fn mat_mul_test_2() {
        let mat1 = vec![
            vec![1.0, 2.0, 3.0],
            vec![4.0, 5.0, 6.0],
            vec![7.0, 8.0, 9.0],
        ];
        let mat2 = vec![vec![1.0], vec![2.0], vec![3.0]];
        let res = mat1.matrix_multiply(&mat2);
        let truth = vec![vec![14.0], vec![32.0], vec![50.0]];
    }

    #[test]
    fn row_transpose_test() {
        let input = vec![1.0, 2.0, 3.0];
        let res = vec![input.clone()].transposed().transposed();

        assert_eq!(vec![input], res)
    }

    #[test]
    fn elem_mult_test() {
        let mat1 = vec![vec![1.0, 2.0, 3.0]];
        let mat2 = vec![vec![3.0, 2.0, 1.0]];

        let res = mat1.transposed().elementwise_multiply(&mat2.transposed());

        assert_eq!(res, vec![vec![3.0, 4.0, 3.0]].transposed());
    }
}