Rust Torch Project

This is a Rust project that uses PyTorch (through tch-rs) for AI and machine learning development.

Project Setup

1. Make sure you have Rust installed. If not, install it using:

   curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

2. Clone this repository:

   git clone <repository-url>
   cd rust-torch

3. The project uses specific versions of dependencies, particularly:

   • tch = "0.19.0" (PyTorch bindings for Rust) Make sure not to change this version as it's compatible with the current libtorch setup.

Installing libtorch on macOS

The good thing is that there's no need to install libtorch manually and globally. Simply run:

make setup

This will automatically:

1. Download the appropriate version of libtorch for macOS
2. Extract it to the correct location within this project
3. Set up the necessary environment variables in the .cargo/config.toml file

That way this project is self-contained, doesn't require any global installations and you don't have to worry about version conflicts.

Building the Project

After setting up libtorch and the environment variables, you can build the project:

cargo build

Project Structure

This project is set up for AI and machine learning development using PyTorch in Rust. It provides a foundation for building:

• Machine Learning models
• Neural Networks
• Deep Learning applications
• AI inference and training pipelines

Troubleshooting

If you encounter any linking errors open an issue on GitHub and I'll help you out.

License

MIT License