Run ruff, isort, pyupgrade, mypy, pylint, flake8, black, blacken-docs, and more on Jupyter Notebooks ? handles IPython magics robustly ? respects your config files ? preserves "quiet mode" trailing semicolons ? lints both code and markdown cells Table of contents Table of contents ?? Installation ?? Examples Command-line Pre-commit ?? Used by ?? Testimonials ?? Contributing ?? Installation In your virtual environment , run (note: the $ is not part of the command): $ python -m pip install -U nbqa To also install all supported linters/formatters: $ python -m pip install -U " nbqa[toolchain] " ?? Examples Command-line Reformat your notebooks with black : $ nbqa black my_notebook.ipynb reformatted my_notebook.ipynb All done! ? ?? ? 1 files reformatted. Sort your imports with isort : $ nbqa isort my_notebook.ipynb --float-to-top Fixing my_notebook.ipynb Upgrade your syntax with pyupgrade : $ nbqa pyupgrade my_notebook.ipynb --py37-plus Rewriting my_notebook.ipynb Format your markdown cells with blacken-docs : $ nbqa blacken-docs my_notebook.ipynb --nbqa-md --nbqa-diff Cell 2 ------ --- my_notebook.ipynb +++ my_notebook.ipynb @@ -1,2 +1 @@ -First level heading -=== +# First level heading To apply these changes, remove the `--nbqa-diff` flag Format .md files saved via Jupytext (requires jupytext to be installed): $ nbqa black my_notebook.md reformatted my_notebook.md All done! ? ?? ? 1 files reformatted. See command-line examples for examples involving doctest , flake8 , mypy , pylint , autopep8 , pydocstyle , yapf , and ruff . Pre-commit Here's an example of how to set up some pre-commit hooks: put this in your .pre-commit-config.yaml file (see usage as pre-commit hook ) - repo : https://github.com/nbQA-dev/nbQA rev : 1.8.5 hooks : - id : nbqa-black additional_dependencies : [jupytext] # optional, only if you're using Jupytext - id : nbqa-pyupgrade args : ["--py37-plus"] - id : nbqa-isort args : ["--float-to-top"] If you need to select specific versions of these linters/formatters, add them to additional_dependencies . ?? Used by Click here for (non-exhaustive) list of repos https://github.com/ComPWA/ampform https://github.com/ComPWA/tensorwaves https://github.com/DataS-DHSC/os-maps-example https://github.com/Eazhi/World-Happiness-Analysis https://github.com/GoogleCloudPlatform/ai-platform-samples https://github.com/MRCIEU/epigraphdb https://github.com/OpenMined/PyDP https://github.com/PlasmaPy/PlasmaPy https://github.com/ProjectPythia/pythia-foundations https://github.com/SeldonIO/alibi https://github.com/TeoZosa/deep-learning-v2-pytorch https://github.com/WM-SEMERU/mlproj_template https://github.com/alan-turing-institute/sktime https://github.com/amor71/LiuAlgoTrader https://github.com/bbernst/base_envs https://github.com/cheginit/HyRiver-examples https://github.com/cheginit/pygeohydro https://github.com/cheginit/pynhd https://github.com/covid-19-impact-lab/sid-germany https://github.com/covid-19-impact-lab/sid https://github.com/dapperfu/Python-Simulink https://github.com/deepcharles/ruptures https://github.com/dhassault/cyclegan_pytorch https://github.com/gboeing/osmnx https://github.com/glotzerlab/freud-examples https://github.com/glotzerlab/hoomd-examples https://github.com/glotzerlab/signac-docs https://github.com/glotzerlab/signac-examples https://github.com/grapl-security/grapl https://github.com/hainegroup/oceanspy https://github.com/henryiii/python-compiled-minicourse https://github.com/intake/intake-esm https://github.com/jameslamb/lightgbm-dask-testing https://github.com/jdb78/pytorch-forecasting https://github.com/jhrcook/advent-of-code_2020 https://github.com/julian-west/asset_price_correlations https://github.com/kratsg/drstorage https://github.com/madebr/pyOpt https://github.com/matthewfeickert/heputils https://github.com/matyama/deep-rl-hands-on https://github.com/mayou36/raredecay https://github.com/neomatrix369/nlp_profiler https://github.com/openforcefield/openff-system https://github.com/pandas-profiling/pandas-profiling https://github.com/paw-lu/dotfiles https://github.com/pawamoy/wps-light https://github.com/phinate/clarinet https://github.com/pik-primap/climate_categories https://github.com/pik-primap/primap2 https://github.com/pik-primap/unfccc_di_api https://github.com/pymc-devs/pymc-examples https://github.com/pymc-devs/resources https://github.com/saturncloud/examples https://github.com/scikit-hep/boost-histogram https://github.com/scikit-hep/iminuit https://github.com/scikit-hep/mplhep https://github.com/scikit-hep/pyhf https://github.com/scikit-hep/pylhe https://github.com/v-goncharenko/data-science-template https://github.com/wemake-services/wemake-python-styleguide https://github.com/zfit/phasespace https://github.com/zfit/zfit https://github.com/zfit/zfit-physics https://github.com/zfit/zfit-tutorials Is your project missing? Let us know, or open a pull request! ?? Testimonials Michael Kennedy & Brian Okken , hosts of the Python Bytes podcast : This is really cool. I think it brings so much of the code formatting and code analysis, clean up to notebooks, which I think had been really lacking Nikita Sobolev , CTO at wemake.services : It is amazing! Alex Andorra , Data Scientist, ArviZ & PyMC Dev, Host of 'Learning Bayesian Statistics' Podcast : well done on nbqa @MarcoGorelli ! Will be super useful in CI Matthew Feickert , Postdoc at University of Illinois working on LHC physics : nbqa in your pre-commit hooks along with @codewithanthony 's pre-commit CI service is amazing! Everyone using Jupyter notebooks should be doing this. Girish Pasupathy , Software engineer and now core-contributor : thanks a lot for your effort to create such a useful tool Simon Brugman , Data scientist & pandas-profiling dev : nbQA helps us to keep notebooks to the same standards as the rest of the code. If you're serious about your code standards, you should keep them consistent across both notebooks and python scripts. Great addition to the ecosystem, thanks! Bradley Dice , PhD Candidate in Physics & Scientific Computing : nbqa is a clean, easy to use, and effective tool for notebook code style. Formatting and readability makes a huge difference when rendering notebooks in a project's documentation! James Lamb , engineer @saturn_cloud, LightGBM maintainer today I learned about nbqa , a command-line tool to run linters like flake8 over #Python code in @ProjectJupyter notebooks. Thanks to @jayyqi for pointing me to it. So far, I really really like it. Lars Yencken , Tech Lead @ Our World In Data Super useful! I only wish it was built-in to Jupyterlab. Vincent D. Warmerdam , maintainer @ calmcode.io Nice. nbQA looks like a great way to prevent the Untitled12.ipynb-phenomenon. I like! Mani Sarkar , Kaggle 3x expert, contributor nbQA is an inspiring project, and the team behind it are very knowledgeable and think of things outside the box with ideas and solutions ?? Contributing I will give write-access to anyone who makes a useful pull request - see the contributing guide for details on how to do so. Thanks goes to these wonderful people ( emoji key ): Marco Gorelli ?? ?? ?? ?? ?? Sebastian Weigand ?? ?? ?? ?? Girish Pasupathy ?? ?? ?? ?? ?? fcatus ?? HD23me ?? mani ?? ?? Daniel Mietchen ?? Michał Gacka ?? Happy ?? Nat Taylor ?? ?? ?? ?? Caio Ariede ?? Nikita Sobolev ?? ?? ?? Amichay Oren ?? pylang ?? Henry Schreiner ?? Kaiqi Dong ?? Simon Brugman ?? John Sandall ?? Nathan Cooper ?? agruenberger ?? Rafal Wojdyla ?? Bradley Dice ?? ?? Ivan Cheung ?? Tony Hirst ?? Taneli Hukkinen ?? Tom Begley ?? ?? ?? Steven DeMartini ?? vincent d warmerdam ? Daniel Sparing ?? asdfCYBER ?? Chris Knight ?? Laura Richter ?? Francesco Ballarin ?? Janosh Riebesell ?? ?? Nicolas Oliver ?? Bas Nijholt ?? Ralf Gommers ?? Toon Verstraelen ?? Trinh Quoc Anh ?? Martin K. Scherer ?? Felix Williams ?? ?? ?? Florian Bar ?? Michael Kennedy ?? Brian Okken ?? Sydney ?? fvonbergen ?? Gaetan Lepage ?? This project follows the all-contributors specification. Contributions of any kind welcome!