We want to make contributing to Opacus is as easy and transparent as possible.
To get the development installation with all the necessary dependencies for linting, testing, and building the documentation, run the following:
git clone https://github.com/pytorch/opacus.git
cd opacus
pip install -e .[dev]
Opacus uses the black and flake8 code formatter to
enforce a common code style across the code base. black is installed easily via
pip using pip install black
, and run locally by calling
black .
flake8 --config ./.circleci/flake8_config.ini
from the repository root. No additional configuration should be needed (see the black documentation for advanced usage).
Opacus also uses isort to sort imports
alphabetically and separate into sections. isort is installed easily via
pip using pip install isort
, and run locally by calling
isort -v -l 88 -o opacus --lines-after-imports 2 -m 3 --trailing-comma .
from the repository root. Configuration for isort is located in .isort.cfg.
We feel strongly that having a consistent code style is extremely important, so CircleCI will fail on your PR if it does not adhere to the black or flake8 formatting style or isort import ordering.
Opacus is fully typed using python 3.6+ type hints. We expect any contributions to also use proper type annotations. While we currently do not enforce full consistency of these in our continuous integration test, you should strive to type check your code locally. For this we recommend using mypy.
To run the unit tests, you can either use pytest
(if installed):
pytest -ra
or python's unittest
:
python -m unittest
To get coverage reports we recommend using the pytest-cov
plugin:
pytest -ra --cov=. --cov-report term-missing
Opacus uses doctest
to ensure our docstrings stay up-to-date. To run all unit tests, including the doctests, run:
python -m pytest --doctest-modules -p conftest opacus
Opacus's website is also open source, and is part of this very repository (the code can be found in the website folder). It is built using Docusaurus, and consists of three main elements:
- The documentation in Docusaurus itself (if you know Markdown, you can already contribute!). This lives in the docs.
- The API reference, auto-generated from the docstrings using Sphinx, and embedded into the Docusaurus website. The sphinx .rst source files for this live in sphinx/source.
- The Jupyter notebook tutorials, parsed by
nbconvert
, and embedded into the Docusaurus website. These live in tutorials.
To build the documentation you will need Node >= 8.x and Yarn >= 1.5.
Run following command from website
folder. It will build the docs and serve the site locally:
./scripts/build_website.sh
You can also perform spell checks on documentation automatically (besides IDEs) using sphinxcontrib-spelling
Note that you will also need PyEnchant
to run sphinxcontrib-spelling
, and thus the Enchant C library. Use this guide for PyEnchant
.
Steps:
- Install the extension with pip:
pip install sphinxcontrib-spelling
- Add
sphinxcontrib.spelling
to the extensions list inconf.py
. - Install
PyEnchant
. Please follow the installation guide. Noticed that Apple Silicons may require a way around under section "Apple Silicon related errors". - Make sure you have a
source
andbuild
folder. Pass "spelling" as the builder argument tosphinx-build
.cd website/sphnix mkdir build # if you do not already have one sphinx-build -b spelling source build
- Find files with spelling errors in
build
(remember to check each folder). A file will be generated for each source file that contains spelling error. Example:- File name:
batch_memory_manager.spelling
- File content:
../../opacus/utils/batch_memory_manager.py:docstring of opacus.utils.batch_memory_manager.BatchMemoryManager:5: (occasinal) safeguarding against occasinal large batches produced by ../../opacus/utils/batch_memory_manager.py:docstring of opacus.utils.batch_memory_manager.BatchMemoryManager:13: (optimzer) On every step optimzer will check if the batch was the last physical batch comprising ../../opacus/utils/batch_memory_manager.py:docstring of opacus.utils.batch_memory_manager.BatchMemoryManager:14: (behaviour) a logical one, and will change behaviour accordignly. ../../opacus/utils/batch_memory_manager.py:docstring of opacus.utils.batch_memory_manager.BatchMemoryManager:14: (accordignly) a logical one, and will change behaviour accordignly. ../../opacus/utils/batch_memory_manager.py:docstring of opacus.utils.batch_memory_manager.BatchSplittingSampler:4: (physocal) Used to split large logical batches into physocal batches of a smaller size,
- File name:
- Manually review the spelling files and make changes in source files accordingly. Some detections are not perfect. For example, "nn" (from torch.nn) can be detected as a spelling error.
We actively welcome your pull requests.
- Fork the repo and create your branch from
main
. - If you have added code that should be tested, add unit tests. In other words, add unit tests.
- If you have changed APIs, document the API change in the PR. Also update the documentation and make sure the documentation builds.
- Ensure the test suite passes.
- Make sure your code passes both
black
andflake8
formatting checks.
We use GitHub issues to track public bugs. Please ensure your description is clear and has sufficient instructions to be able to reproduce the issue.
Facebook has a bounty program for the safe disclosure of security bugs. In those cases, please go through the process outlined on that page and do not file a public issue.
By contributing to Opacus, you agree that your contributions will be licensed under the LICENSE file in the root directory of this source tree.