vinyana 0.1.0

A 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
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
158
159
160
161
162
163
164
165

use std::{error::Error, fs::File, io::Read, path::Path};

use matrix::Matrix;
use rand::{thread_rng, Rng};
use serde::{Deserialize, Serialize};

mod matrix;
pub const E: f32 = 2.7182818284590451f32;

pub fn sigmoid(x: f32) -> f32 {
    1.0 / (1.0 + E.powf(-x))
}

pub fn sigmoid_derivate(x: f32) -> f32 {
    x * (1.0 - x)
}

pub fn relu(x: f32) -> f32 {
    x.max(0.0)
}

pub fn tahn(x: f32) -> f32 {
    (E.powf(x) - E.powf(-x)) / (E.powf(x) + E.powf(-x))
}

/// First iteration of Neural Network with only 3 layers input -> hidden -> output
#[derive(Serialize, Deserialize)]
pub struct NeuralNetwork {
    pub input_nodes: usize,
    pub hidden_nodes: usize,
    pub output_nodes: usize,
    pub learning_rate: f32,
    pub w_input_hidden: Matrix,
    pub w_hidden_output: Matrix,
    pub w_bias_hidden: Matrix,
    pub w_bias_output: Matrix,
}

impl NeuralNetwork {
    /// The posibility to initialise a completly new neural network
    /// All the weights will be random from -1 to 1 floats
    pub fn new(
        input_nodes: usize,
        hidden_nodes: usize,
        output_nodes: usize,
        learning_rate: f32,
    ) -> NeuralNetwork {
        let rng = thread_rng();

        let mut w_input_hidden = Matrix::new(hidden_nodes, input_nodes);
        let mut w_hidden_output = Matrix::new(output_nodes, hidden_nodes);

        let mut w_bias_hidden = Matrix::new(hidden_nodes, 1);
        let mut w_bias_output = Matrix::new(output_nodes, 1);

        let randomize = |_| rng.clone().gen_range(-1.0..1.0);

        // randomize weights
        w_input_hidden = w_input_hidden.map(randomize);
        w_hidden_output = w_hidden_output.map(randomize);

        // randomize biases
        w_bias_hidden = w_bias_hidden.map(randomize);
        w_bias_output = w_bias_output.map(randomize);

        NeuralNetwork {
            input_nodes,
            hidden_nodes,
            output_nodes,
            learning_rate,
            w_input_hidden,
            w_hidden_output,
            w_bias_hidden,
            w_bias_output,
        }
    }

    pub fn load(filepath: &Path) -> Result<NeuralNetwork, Box<dyn Error>> {
        let mut file = File::open(filepath)?;

        let mut contents = vec![];
        file.read_to_end(&mut contents)?;

        let nn: NeuralNetwork = bincode::deserialize(&contents)?;

        Ok(nn)
    }

    pub fn save(&self, filepath: &Path) -> Result<(), Box<dyn Error>> {
        let encoded = bincode::serialize(self).expect("Could not serialize the model");
        std::fs::write(filepath, encoded).expect("Could not save the model in the provided path");

        Ok(())
    }

    pub fn train(&mut self, inputs: &Vec<f32>, targets: &Vec<f32>) {
        // Do the prediction logic
        let inputs = Matrix::from_vec(inputs);

        let hidden_inputs = &(&self.w_input_hidden * &inputs) + &self.w_bias_hidden;
        let hidden_outputs = hidden_inputs.map(sigmoid);

        let final_inputs = &(&self.w_hidden_output * &hidden_outputs) + &self.w_bias_output;
        let final_outputs = final_inputs.map(sigmoid);

        let targets = Matrix::from_vec(targets);
        let output_errors = &targets - &final_outputs;

        // Compute the output gradient
        let output_gradients = final_outputs.map(sigmoid_derivate);
        let output_gradients = output_gradients.mul(&output_errors);
        let output_gradients = &output_gradients * self.learning_rate;

        let w_hidden_output_deltas = &output_gradients * &(hidden_outputs.transpose());

        self.w_hidden_output = &self.w_hidden_output + &w_hidden_output_deltas;
        self.w_bias_output = &self.w_bias_output + &output_gradients;

        // compute the hidden layer errors

        let w_hidden_output_transposed = self.w_hidden_output.transpose();
        let hidden_errors = &w_hidden_output_transposed * &output_errors;

        let hidden_gradients = hidden_outputs.map(sigmoid_derivate);
        let hidden_gradients = hidden_gradients.mul(&hidden_errors);
        let hidden_gradients = &hidden_gradients * self.learning_rate;

        let inputs_transposed = inputs.transpose();
        let w_input_hidden_deltas = &hidden_gradients * &inputs_transposed;

        self.w_input_hidden = &self.w_input_hidden + &w_input_hidden_deltas;
        self.w_bias_hidden = &self.w_bias_hidden + &hidden_gradients;
    }
    pub fn predict(&self, inputs: &Vec<f32>) -> Vec<f32> {
        let inputs = Matrix::from_vec(inputs);

        let hidden_inputs = &(&self.w_input_hidden * &inputs) + &self.w_bias_hidden;
        let hidden_outputs = hidden_inputs.map(sigmoid);

        let final_inputs = &(&self.w_hidden_output * &hidden_outputs) + &self.w_bias_output;
        let final_outputs = final_inputs.map(sigmoid);

        final_outputs
            .data
            .iter()
            .flatten()
            .map(|x| *x)
            .collect::<Vec<_>>()
    }
}

#[cfg(test)]
mod tests {

    use crate::NeuralNetwork;

    #[test]
    fn test_neural_network_initialisation() {
        let nn = NeuralNetwork::new(3, 3, 1, 0.001);

        assert_eq!(nn.hidden_nodes, 3);
        assert_eq!(nn.input_nodes, 3);
        assert_eq!(nn.output_nodes, 1);
    }
}