Neural Network in C++

A simple neural network implementation in C++. This project is a learning exercise to improve C++ skills and understand neural network basics.

Features

• Tensor class for matrix operations
• torch.nn like modules like Linear, ReLU
• Forward and backward pass using optimizer

Usage

1. Clone the repository:

   git clone github.com/gladuz/nn-cpp.git
   

2. Compile:

   make
   

3. Run:

   ./build/nn_project
   

Verification with toy example

MNIST example digits printed on terminal

Training code

Tensor data({1, 2, 3, 4, 5, 6}, 2, 3);  // 2x3 matrix
Tensor target({2, -2}, 2, 1);

nn::Sequential seq;
seq.add(std::make_shared<nn::Linear>(3, 5));
seq.add(std::make_shared<nn::ReLU>());
seq.add(std::make_shared<nn::Linear>(5, 7));
seq.add(std::make_shared<nn::ReLU>());
seq.add(std::make_shared<nn::Linear>(7, 1));
auto optimizer = optim::SGD(seq.parameters(), 0.01);

for(int i=0; i<100; i++){
   auto out = seq.forward(data);
   auto [loss, loss_grad] = nn::functional::mse_loss(out, target);
   cout<<"Loss at epoch: "<<i<<" is: "<<loss<<endl;
   seq.backward(loss_grad);
   optimizer.step();
}

Future Plans

• Add dataloader example with MNIST
• Implement additional layer types