astrai 2.2.0

A pretty bad neural network 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

use super::*;

#[derive(Debug, Clone, Copy, Default, Hash, PartialEq, Eq, PartialOrd, Ord, EnumIter)]
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
pub enum ActivationFunction {
    #[default]
    Sigmoid,
    SigmoidDerivative,
    Relu,
    Tanh,
    TanhDerivative,
    BinaryStep,
    Identity,
}

impl ActivationFunction {
    pub fn call(&self, x: f64) -> f64 {
        match self {
            ActivationFunction::Sigmoid => sigmoid(x),
            ActivationFunction::SigmoidDerivative => sigmoid_derivative(x),
            ActivationFunction::Relu => relu(x),
            ActivationFunction::Tanh => tanh(x),
            ActivationFunction::TanhDerivative => tanh_derivative(x),
            ActivationFunction::BinaryStep => binary_step(x),
            ActivationFunction::Identity => x,
        }
    }

    pub fn derivative(&self, x: f64) -> f64 {
        match self {
            ActivationFunction::Sigmoid => sigmoid_derivative(x),
            ActivationFunction::SigmoidDerivative => sigmoid_derivative(x),
            ActivationFunction::Relu => 1.0,
            ActivationFunction::Tanh => tanh_derivative(x),
            ActivationFunction::TanhDerivative => tanh_derivative(x),
            ActivationFunction::BinaryStep => 0.0,
            ActivationFunction::Identity => 1.0,
        }
    }
}

pub fn sigmoid(x: f64) -> f64 {
    1.0 / (1.0 + (-x).exp())
}

pub fn sigmoid_derivative(x: f64) -> f64 {
    let s = sigmoid(x);
    s * (1.0 - s)
}

pub fn relu(x: f64) -> f64 {
    if x > 0.0 {
        x
    } else {
        0.0
    }
}

pub fn tanh(x: f64) -> f64 {
    x.tanh()
}

pub fn tanh_derivative(x: f64) -> f64 {
    let t = tanh(x);
    1.0 - t.powi(2)
}

pub fn binary_step(x: f64) -> f64 {
    if x > 0.0 {
        1.0
    } else {
        0.0
    }
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_sigmoid() {
        assert_eq!(sigmoid(0.0).round(), (0.5f64).round());
        assert_eq!(sigmoid(1.0).round(), (0.7310585786300049f64).round());
        assert_eq!(sigmoid(-1.0).round(), (0.2689414213699951f64).round());
    }
    #[test]
    fn test_sigmoid_derivative() {
        assert_eq!(sigmoid_derivative(0.0).round(), 0.25f64.round());
        assert_eq!(
            sigmoid_derivative(1.0).round(),
            0.19661193324148185f64.round()
        );
        assert_eq!(
            sigmoid_derivative(-1.0).round(),
            0.19661193324148185f64.round()
        );
    }

    #[test]
    fn test_relu() {
        assert_eq!(relu(0.0).round(), 0f64.round());
        assert_eq!(relu(1.0).round(), 1f64.round());
        assert_eq!(relu(-1.0).round(), 0f64.round());
    }

    #[test]
    fn test_tanh() {
        assert_eq!(tanh(0.0).round(), 0.0f64.round());
        assert_eq!(tanh(1.0).round(), 0.7615941559557649f64.round());
        assert_eq!(tanh(-1.0).round(), -0.7615941559557649f64.round());
    }

    #[test]
    fn test_tanh_derivative() {
        assert_eq!(tanh_derivative(0.0).round(), 1.0f64.round());
        assert_eq!(tanh_derivative(1.0).round(), 0.41997434161402614f64.round());
        assert_eq!(
            tanh_derivative(-1.0).round(),
            0.41997434161402614f64.round()
        );
    }

    #[test]
    fn test_binary_step() {
        assert_eq!(binary_step(0.0).round(), 0f64.round());
        assert_eq!(binary_step(1.0).round(), 1f64.round());
        assert_eq!(binary_step(-1.0).round(), 0f64.round());
    }
}