Runtime Entrypoint

This is the broad import for TorchLean's executable layer. It collects the runtime pieces that are useful when building, training, importing, exporting, or checking runnable models:

• the eager and compiled autograd engines;
• the lower-level Runtime.Autograd.Torch session operations;
• the higher-level Runtime.Autograd.TorchLean front-end used by NN.API.Runtime;
• deterministic dataset/training utilities;
• optional external-process helpers for untrusted producer / trusted checker workflows;
• pure optimizer and scheduler equations;
• PyTorch import/export bridge infrastructure; and
• typed reinforcement-learning runtime helpers.

For ordinary user code, prefer import NN or import NN.API.Runtime. Import this file when you intentionally want the full executable subsystem. If you only need pure tensor semantics and theorems, prefer NN.Entrypoint.Spec or NN.Entrypoint.Proofs; those imports keep runtime bridge dependencies out of the build.

The runtime entrypoint deliberately imports only reusable bridge infrastructure under NN.Runtime.PyTorch.*. Demo-only MLP/CNN/Transformer round-trip code lives under NN.Examples.Interop.PyTorch.*, so ordinary runtime imports do not pull example modules into the library surface.

References / context:

• PyTorch autograd overview: https://pytorch.org/docs/stable/autograd.html
• PyTorch nn.Module / tensor ops surface: https://pytorch.org/docs/stable/nn.html https://pytorch.org/docs/stable/torch.html
• TorchLean’s import/export bridge details live in NN.Runtime.PyTorch.Export.Core and NN.Runtime.PyTorch.Import.Core.