pytorch/torch/csrc/jit
Mike Ruberry 56af122659 Revert D23966878: [pytorch][PR] This PR flips a switch to enable PE + TE
Test Plan: revert-hammer

Differential Revision:
D23966878 (dddb685c11)

Original commit changeset: 2010a0b07c59

fbshipit-source-id: 132556039730fd3e4babd0d7ca8daf9c8d14f728
2020-09-29 04:33:19 -07:00
..
api [JIT] Add property support to TorchScript classes (#42389) 2020-08-14 12:56:57 -07:00
backends [JIT] Modify to_backend API so that it accepts wrapped modules (#43612) 2020-09-28 17:17:01 -07:00
codegen [nvFuser] Latency improvements for pointwise + reduction fusion (#45218) 2020-09-24 23:17:20 -07:00
docs [jit] update to serialization doc (#39025) 2020-06-05 17:49:08 -07:00
frontend Trace scattered tensor options arguments (#44071) 2020-09-25 09:04:06 -07:00
ir [JIT] improve alias analysis for list constructs (#39111) 2020-09-22 09:38:59 -07:00
mobile log metadata when model loading failed (#44430) 2020-09-24 20:09:22 -07:00
passes Revert D23966878: [pytorch][PR] This PR flips a switch to enable PE + TE 2020-09-29 04:33:19 -07:00
python [ONNX] Optimize export_onnx api to reduce string and model proto exchange (#44332) 2020-09-27 16:29:08 -07:00
runtime Revert D23966878: [pytorch][PR] This PR flips a switch to enable PE + TE 2020-09-29 04:33:19 -07:00
serialization [ONNX] Optimize export_onnx api to reduce string and model proto exchange (#44332) 2020-09-27 16:29:08 -07:00
tensorexpr [wip] fast typeMeta/ScalarType conversion approach 2 (#44965) 2020-09-29 02:39:36 -07:00
testing Fix check highlight in filecheck. (#42417) 2020-08-03 18:49:22 -07:00
jit_log.cpp [jit] Speed up saving in case of many classes (#44589) 2020-09-15 15:10:45 -07:00
jit_log.h make sure logs work inside aten/c10 namespaces as well (#37018) 2020-04-21 20:01:13 -07:00
OVERVIEW.md Add documentation for PYTORCH_JIT_TYPE_VERBOSITY (#42241) 2020-08-06 10:39:39 -07:00
README.md
resource_guard.h

PyTorch JIT

This folder contains (most of) the C++ code for the PyTorch JIT, a language and compiler stack for executing PyTorch models portably and efficiently. To learn more about the JIT from a user perspective, please consult our reference documentation and tutorials.

A brief summary of the source tree:

  • OVERVIEW.md: High-level technical overview of the JIT.
  • frontend/: Taking PyTorch modules in Python and translating them into the JIT IR.
  • ir/: Core IR abstractions.
  • runtime/: Interpreter, graph execution, and JIT operators.
  • codegen/: Generating efficient, hardware-specific code for JIT subgraphs.
  • serialization/: Saving and loading modules.
  • api/: Any user-facing C++ or Python interfaces.
  • python/: Binding stuff into Python or accessing information from the Python environment.
  • testing/: Utilities and helpers for testing.
  • mobile/: Mobile-specific implementations of runtime components.
  • passes/: IR-to-IR passes, generally for optimization and lowering.
  • generated/: This folder is generated by the PyTorch build, and contains bindings for native PyTorch operators into the JIT.

Refer to each folder for more in-depth documentation.

Other relevant parts of the codebase not contained here:

  • aten/src/ATen/core: contains JIT code re-used by other elements of the runtime system (eager, mobile, etc.)