pytorch/tools
albanD b8895df8db Fix black binary again for debug python (#86275)
The `--no-binary` flag was not ported when moving from black only to ufmt.
This adds it back.

This is to work around the fact that black binary hard crashes when running with debug python and it needs to be compiled from source.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/86275
Approved by: https://github.com/bdhirsh, https://github.com/malfet
2022-10-05 21:08:40 +00:00
..
amd_build [ROCm] enable nvfuser (#82498) 2022-08-30 21:50:39 +00:00
autograd Revert "Revert "SymIntify cat and narrow (#86191)"" (#86289) 2022-10-05 20:51:28 +00:00
bazel_tools
build_defs
code_analyzer
code_coverage
config
coverage_plugins_package
fast_nvcc
gdb
iwyu
jit
linter Fix black binary again for debug python (#86275) 2022-10-05 21:08:40 +00:00
lite_interpreter
lldb
onnx [ONNX] Initial version of diagnostics infrastructure. (#85107) 2022-09-30 07:47:26 +00:00
pyi improve annotations (#86105) 2022-10-05 10:33:26 +00:00
rules
setup_helpers [Mobile] Fix The Build For Model Tracer (#84755) 2022-09-09 18:22:24 +00:00
shared
stats
test fix external codegen kernel error checking (#85029) 2022-09-17 04:08:09 +00:00
testing parallelize at file granularity (#85770) 2022-10-03 16:59:39 +00:00
__init__.py
bazel.bzl
BUCK.bzl Add torch.nested namespace (#84102) 2022-09-12 16:31:05 +00:00
BUCK.oss
build_libtorch.py
build_pytorch_libs.py
download_mnist.py
extract_scripts.py
gen_flatbuffers.sh
gen_vulkan_spv.py
generate_torch_version.py
generated_dirs.txt
git_add_generated_dirs.sh
git_reset_generated_dirs.sh
nightly.py
nvcc_fix_deps.py
pytorch.version
README.md
render_junit.py
substitute.py
update_masked_docs.py [maskedtensor] port torch/_masked into torch/masked (#85515) 2022-09-26 23:41:13 +00:00
vscode_settings.py

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: