Modular neural network in Rust.

Create modular neural networks in Rust with ease! For educational purposes; vector operations are not throughly optimized.

Quickstart

use neurons::network::Network;
use neurons::activation::Activation;
use neurons::optimizer::Optimizer;
use neurons::objective::Objective;

fn main() {
    let mut network = Network::new();

    network.add_layer(4, 50, activation::Activation::Linear, true);
    network.add_layer(50, 50, activation::Activation::Linear, true);
    network.add_layer(50, 1, activation::Activation::Linear, false);
    
    network.set_optimizer(
        optimizer::Optimizer::AdamW(
            optimizer::AdamW {
                learning_rate: 0.001,
                beta1: 0.9,
                beta2: 0.999,
                epsilon: 1e-8,
                decay: 0.01,

                momentum: vec![],           // To be filled by the network
                velocity: vec![],           // To be filled by the network
            }
        )
    );
    network.set_objective(
        objective::Objective::MSE,          // Objective function
        Some((-1f32, 1f32))                 // Gradient clipping
    );
  
    println!("{}", network);
  
    let (x, y) = ...;                       // Load data
    let epochs = 1000;
    let loss = network.learn(x, y, epochs); // Train the network
}

Examples can be found in the examples directory.

Progress

• Layer types

  • Dense
  • Feedback
  • Convolutional

• Activation functions

  • Linear
  • Sigmoid
  • Tanh
  • ReLU
  • LeakyReLU
  • Softmax

• Objective functions

  • AE
  • MAE
  • MSE
  • RMSE
  • CrossEntropy
  • BinaryCrossEntropy
  • KLDivergence
  • Huber

• Optimization techniques

  • SGD
  • SGDM
  • Adam
  • AdamW
  • RMSprop
  • Minibatch

• Regularization

  • Dropout
  • Batch normalization
  • Early stopping

• Parallelization

  • Multi-threading

• Testing

  • Unit tests
  • Integration tests

• Other

  • Documentation
  • Type conversion (e.g. f32, f64)
  • Network type specification (e.g. f32, f64)
  • Saving and loading
  • Logging
  • Data from file
  • Custom tensor/matrix types
  • Custom random weight initialization
  • Plotting

Inspiration

• candle
• rust-simple-nn

Sources

• backpropagation

• softmax

• momentum

• Adam

• AdamW

• RMSprop

• GitHub Copilot

• ChatGPT

• Mistral