Shadowgraph

Tech stack & purpose

Ocaml-torch provides OCaml bindings for PyTorch, enabling machine learning and deep learning work in the OCaml language. The project spans multiple problem domains within machine learning, including neural network training, generative models, computer vision tasks, and model inference. The tech stack centers on OCaml as the host language, with bindings to PyTorch for tensor operations and neural network construction, plus integration with related libraries like torch_vision for image processing. Additional tools mentioned include dune (the OCaml build system) and Torch Script for model serialization and cross-language deployment.