saymrwulf/pytorch
1
0
Fork 0
mirror of https://github.com/saymrwulf/pytorch.git synced 2026-05-14 20:57:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
pytorch/tools
History
Vitaly Fedyunin 4bfe2f0900 Fix jit outplace tracing and reapply changes to *_like operators. (#28839) 
Summary:
Reapply reverted and fix files `gen_variable_type.py` `test_jit.py`

https://github.com/pytorch/pytorch/issues/27891 Cleanup testing of _like operators
https://github.com/pytorch/pytorch/issues/27890 Add memory format support to randn_like operator
https://github.com/pytorch/pytorch/issues/27889 Add memory format support to randint_like operator
https://github.com/pytorch/pytorch/issues/27562 Add memory format support to zeros_like operator
https://github.com/pytorch/pytorch/issues/27561 Add memory format support to rand_like operator
https://github.com/pytorch/pytorch/issues/27270 Add memory format support to ones_like operator
https://github.com/pytorch/pytorch/issues/27262 Add memory format support to full_like operator
Pull Request resolved: https://github.com/pytorch/pytorch/pull/28839

Test Plan:
Imported from GitHub, without a `Test Plan:` line.

buck test mode/dev //language_technology/neural_mt/os/pytorch_translate/test:test_onnx -- 'test_forced_decoder_export_vocab_reduction \(language_technology\.neural_mt\.os\.pytorch_translate\.test\.test_onnx\.TestONNX\)'

Differential Revision: D18203397

Pulled By: VitalyFedyunin

fbshipit-source-id: eea41cbd4c232cf5a54172b1e1b16b173798f298
2019-10-31 13:23:08 -07:00
..
amd_build Enable EXE001 flake8 check. (#27560) 2019-10-09 09:15:29 -07:00
autograd Fix jit outplace tracing and reapply changes to *_like operators. (#28839) 2019-10-31 13:23:08 -07:00
docker clean up runtime dockerfile, use cuda 9 package (#7230) 2018-05-02 23:54:05 -07:00
jit Migrate return type void to () for native functions. (#28290) 2019-10-22 15:23:20 -07:00
pyi Improve Tensor type hints (#28578) 2019-10-27 04:43:51 -07:00
setup_helpers Remove tools/setup_helpers/nvtoolext.py (do not seem to be used) (#28125) 2019-10-17 09:07:09 -07:00
shared Ignore F401 in all __init__.py without putting noqa (#25823) 2019-10-23 15:28:13 -07:00
__init__.py python 2 support 2016-06-08 19:14:57 -04:00
aten_mirror.sh Restore TBB module (#20454) 2019-05-28 02:49:36 -07:00
build_libtorch.py Specify build dir as a global variable in BUILD_DIR in the build system. 2019-07-25 07:19:47 -07:00
build_pytorch_libs.py remove tools/setup_helpers/cudnn.py (#25876) 2019-09-24 07:44:33 -07:00
build_variables.py Switching tests to ProfilingExecutor (rebased) 2019-10-29 11:41:42 -07:00
clang_format.py Enable EXE001 flake8 check. (#27560) 2019-10-09 09:15:29 -07:00
clang_tidy.py add clang-tidy to github actions (#27755) 2019-10-11 17:01:50 -07:00
download_mnist.py Turn on F401: Unused import warning. (#18598) 2019-03-30 09:01:17 -07:00
flake8_hook.py Add missing shebangs to Python files with executable permissions. 2019-06-06 10:53:40 -07:00
generated_dirs.txt Add simple scripts for checking if generated code changed. (#12835) 2018-10-22 07:33:32 -07:00
git-pre-commit Remove THD (#22065) 2019-06-25 12:19:13 -07:00
git_add_generated_dirs.sh Add simple scripts for checking if generated code changed. (#12835) 2018-10-22 07:33:32 -07:00
git_reset_generated_dirs.sh Add simple scripts for checking if generated code changed. (#12835) 2018-10-22 07:33:32 -07:00
pytorch.version Convert all tabs to spaces, add CI. (#18959) 2019-04-09 08:12:26 -07:00
README.md Stop doing nn wrap. (#25353) 2019-08-30 07:42:20 -07:00

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.

Legacy infrastructure (we should kill this):

  • cwrap - Implementation of legacy code generation for THNN/THCUNN. This is used by nnwrap.

Build system pieces:

  • setup_helpers - Helper code for searching for third-party dependencies on the user system.
  • build_pytorch_libs.sh - Script that builds all of the constituent libraries of PyTorch, but not the PyTorch Python extension itself. We are working on eliminating this script in favor of a unified cmake build.
  • build_pytorch_libs.bat - Same as above, but for Windows.
  • build_libtorch.py - Script for building libtorch, a standalone C++ library without Python support. This build script is tested in CI.

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: