v2 of https://github.com/pytorch/pytorch/pull/102125 because of git issues
corresponding deserialization diff: https://github.com/pytorch/pytorch/pull/102716
Implementing serialization of the exported program to a python dataclass, and then from that dataclass to json. This is split into a couple of sections:
- `serialize(ep: ep.ExportedProgram, opset_version: Dict[str, int]) -> Tuple[bytes, bytes]` -- takes an exported program object, a dictionary mapping opset namespaces to versions, and returns the serialized exported program in bytes, and separately the state dict serialized in bytes
- `GraphModuleSerializer` class that serializes torch.fx.GraphModule
to the schema.GraphModule dataclass
- `ExportedProgramSerializer` class that serializes torch._export.exported_program.ExportedProgram to the schema.ExportedProgram dataclass
Serialization TODOs:
- [x] pytree spec: https://github.com/pytorch/pytorch/pull/102577
- [ ] higher order ops
- [ ] node metadata (specifically nn_module_stack/source_fn)
- [ ] constraints
- [ ] graph module metadata
The tests are not super comprehensive, but that's because I think it'll be better tested + easier to test once deserialization is implemented.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102707
Approved by: https://github.com/avikchaudhuri, https://github.com/zhxchen17
