Torch Core

Torch-style runtime front-end for eager execution, compiled wrappers, and training helpers.

• Core.Types: public handles, options, and parameter wrappers.
• Core.Session: eager session state, CUDA bridge, and tape lifecycle helpers.
• Core.Ops: eager tensor operations.
• Core.BackwardOptim: eager backward passes and optimizers.
• Core.Compiled: proof-backed compiled wrappers.
• Core.Functional: backend-generic Ops interface and curried syntax.
• Core.Trainer: Ops instances, parameter packs, and scalar trainer construction.

Most callers can import this module directly to bring the Torch runtime API into scope.