Guides#
These guides cover the working surfaces of LEAPP in more depth. Read them in order or jump to the topic you need.
Distributed inputs, method shorthand, and nested data structures.
Backend selection, TorchScript vs ONNX, and bringing pre-compiled models into the graph.
Explicit state tensors, module buffer tracking, and automatic feedback detection for recurrent or re-entered graph state.
Registered module buffers, static outputs, embedded constants, and manual tensor-copy tag preservation.
Dry-run modes, selective non-traced nodes, verbose logs, and full FX graph inspection.
Per-node validation, cached re-entry examples, and the Python
InferenceManager for end-to-end smoke tests.
Use TensorSemantics to describe what tensors represent and name
individual tensor elements.