Google Git
Sign in
android / platform / external / pytorch / 615dd25088 / . / tools
tree: 803d7ec683732a4faea6a326b585285ac298a4ae [path history] [tgz]
  1. amd_build/
  2. autograd/
  3. bazel_tools/
  4. build_defs/
  5. code_analyzer/
  6. code_coverage/
  7. config/
  8. coverage_plugins_package/
  9. fast_nvcc/
  10. gdb/
  11. iwyu/
  12. jit/
  13. linter/
  14. lite_interpreter/
  15. lldb/
  16. onnx/
  17. pyi/
  18. rules/
  19. setup_helpers/
  20. shared/
  21. stats/
  22. test/
  23. testing/
  24. __init__.py
  25. bazel.bzl
  26. build_libtorch.py
  27. build_pytorch_libs.py
  28. cpuinfo_target_definition.bzl
  29. download_mnist.py
  30. extract_scripts.py
  31. gen_flatbuffers.sh
  32. generate_torch_version.py
  33. generated_dirs.txt
  34. git_add_generated_dirs.sh
  35. git_reset_generated_dirs.sh
  36. miniz_target_definition.bzl
  37. nightly.py
  38. nvcc_fix_deps.py
  39. perf_kernel_defs.bzl
  40. pytorch.version
  41. README.md
  42. render_junit.py
  43. sgx_aten_target_definitions.bzl
  44. sgx_caffe2_target_definitions.bzl
  45. sgx_target_definitions.bzl
  46. substitute.py
  47. target_definitions.bzl
  48. update_masked_docs.py
  49. vscode_settings.py
tools/README.md

This folder contains a number of scripts which are used as part of the PyTorch build process. This directory also doubles as a Python module hierarchy (thus the __init__.py).

Overview

Modern infrastructure:

  • autograd - Code generation for autograd. This includes definitions of all our derivatives.
  • jit - Code generation for JIT
  • shared - Generic infrastructure that scripts in tools may find useful.
    • module_loader.py - Makes it easier to import arbitrary Python files in a script, without having to add them to the PYTHONPATH first.

Build system pieces:

  • setup_helpers - Helper code for searching for third-party dependencies on the user system.
  • build_pytorch_libs.py - cross-platform script that builds all of the constituent libraries of PyTorch, but not the PyTorch Python extension itself.
  • build_libtorch.py - Script for building libtorch, a standalone C++ library without Python support. This build script is tested in CI.
  • fast_nvcc - Mostly-transparent wrapper over nvcc that parallelizes compilation when used to build CUDA files for multiple architectures at once.
    • fast_nvcc.py - Python script, entrypoint to the fast nvcc wrapper.

Developer tools which you might find useful:

Important if you want to run on AMD GPU:

  • amd_build - HIPify scripts, for transpiling CUDA into AMD HIP. Right now, PyTorch and Caffe2 share logic for how to do this transpilation, but have separate entry-points for transpiling either PyTorch or Caffe2 code.
    • build_amd.py - Top-level entry point for HIPifying our codebase.

Tools which are only situationally useful: