Contributing to Econ-ARK¶
The Econ-ARK project has a Code of Conduct to which all contributors must adhere.
Thank you for considering contributing to Econ-ARK! We’re a young project with a small but committed community that’s hoping to grow while maintaining our friendly and responsive culture. Whether you’re an economist or a technologist, a writer or a coder, an undergrad or a full professor, a professional or a hobbyist, there’s a place for you in the Econ-ARK community.
We’re still creating our contribution infrastructure, so this document is a work in progress. If you have any questions please feel free to @ or otherwise reach out to project leaders Chris and Matt. If you prefer to connect through email, you can send it to econ-ark at jhuecon dot org. We have a Gitter chat room you are welcome to meet us in, as well as a Discord channel listed under the channel for the Scientific Python community!
How to Contribute¶
We’re open to all kinds of contributions, from bug reports to help with our docs to suggestions on how to improve our code. The best way to figure out if the contribution you’d like to make is something we’d merge or otherwise accept, is to open up an issue in our issue tracker. Please create an issue rather than immediately submitting pull request, unless the change you’d like to make is so minor you won’t mind if the pull request is rejected. For bigger contributions, we want to proactively talk things through so we don’t end up wasting your time.
While we’re thrilled to receive all kinds of contributions, there are a few key areas we’d especially like help with:
porting existing heterogenous agent/agent based models into HARK
curating and expanding the collection of projects which use Econ-ARK (which we store in the remark repository)
creating demonstrations of how to use Econ-ARK (which we store in the DemARK repository)
expanding test coverage of our existing code
If you’d like to help with those or any other kind of contribution, reach out to us and we’ll help you do so.
We don’t currently have guidelines for opening issues or pull requests, so include as much information as seems relevant to you, and we’ll ask you if we need to know more.
Responding to Issues & Pull Requests¶
We’re trying to get better at managing our open issues and pull requests. We’ve created a new set of goals for all issues and pull requests in our Econ-ARK repos:
Initial response within one or two days.
Substantive response within two weeks.
Resolution of issue/pull request within three months.
If you’ve been waiting on us for more than two weeks for any reason, please feel free to give us a nudge. Correspondingly, we ask that you respond to any questions or requests from us within two weeks as well, even if it’s just to say, “Sorry, I can’t get to this for a while yet”. If we don’t hear back from you, we may close your issue or pull request. If you want to re-open it, just ask - we’re glad to do so.
The Contributing Guide below provides instructions for how to get started running HARK. This also serves as a setup guide for new contributors. If you run into any problems, please let us know by opening an issue in the issue tracker.
Thanks again! We’re so glad to have you in our community.
If you are a first-time contributor:
Go to https://github.com/econ-ark/HARK and click the “fork” button to create your own copy of the project.
Clone the project to your local computer
git clone email@example.com:your-username/HARK.git
Navigate to the folder HARK and add the upstream repository
git remote add upstream firstname.lastname@example.org:econ-ark/HARK.git
Now, you have remote repositories named:
upstream, which refers to the
origin, which refers to your personal fork of
Develop your contribution:
Pull the latest changes from upstream
git checkout master git pull upstream master
Create a branch for the feature you want to work on. Since the branch name will appear in the merge message, use a sensible name such as ‘bugfix-for-issue-220’
git checkout -b bugfix-for-issue-220
Commit locally as you progress (
To submit your contribution:
Push your changes back to your fork on GitHub
git push origin bugfix-for-issue-220
Go to GitHub. The new branch will show up with a green Pull Request button—click it.
Reviewers (the other developers and interested community members) will write inline and/or general comments on your Pull Request (PR) to help you improve its implementation, documentation, and style. Every single developer working on the project has their code reviewed, and we’ve come to see it as friendly conversation from which we all learn and the overall code quality benefits. Therefore, please don’t let the review discourage you from contributing: its only aim is to improve the quality of project, not to criticize (we are, after all, very grateful for the time you’re donating!).
To update your pull request, make your changes on your local repository and commit. As soon as those changes are pushed up (to the same branch as before) the pull request will update automatically.
Travis-CI, a continuous integration service, is triggered after each Pull Request update to build the code and run unit tests of your branch. The Travis tests must pass before your PR can be merged. If Travis fails, you can find out why by clicking on the “failed” icon (red cross) and inspecting the build and test log.
GitHub Actions, is another continuous integration service, which we use. You will also need to make sure that the GitHub Actions tests pass.
NOTE: If closing a bug, also add “Fixes #1480” where 1480 is the issue number.
Build environment setup¶
Once you’ve cloned your fork of the HARK repository,
you should set up a Python development environment tailored for HARK.
You may choose the environment manager of your choice.
Here we provide instructions for two popular environment managers:
venv (pip based) and
conda (Anaconda or Miniconda).
venv, you may find the following bash commands useful
# Create a virtualenv named ``econ-dev`` that lives in the directory of # the same name python -m venv econ-dev # Activate it source econ-dev/bin/activate # Build and install HARK from source with developer requirements pip install -e ".[dev]" # Test your installation pip install pytest pytest HARK/
When using conda, you may find the following bash commands useful
# Create a conda environment named ``econ-dev`` conda create --name econ-dev # Activate it conda activate econ-dev # Install minimal testing dependencies conda install pytest # Build and install HARK from source with developer requirements pip install -e ".[dev]" # Test your installation pytest HARK/
All code should have tests.
All code should be documented, to the same standard as NumPy and SciPy.
All changes are reviewed.
black <https://black.readthedocs.io>_ for styling of code
# install black pip install black # run black on the changed files black path_to_changed_file.py
Object naming conventions in HARK are fairly different than existing standards, and differ somewhat between tool modules vs model or application modules. The following conventions apply throughout HARK:
Functions and methods are always in ‘’camel case’’: no underscores, first letter is lower case, first letter of each subsequent word is capitalized. E.g. approxLognormal.
Function and method names should accurately and concisely describe what the function does; the first word in the name must be a verb
Variable and class names should not have a verb as their first word.
Class names should use no underscores and capitalize the first letter of each word; moreover, a class name must include a noun. E.g. ConsPerfForesightSolver.
When naming variables in model modules, the HARK team strongly discourages using single letter names, like c for consumption. Instead, we encourage contributors to use longer, more descriptive variable names using additional words and common abbreviations to specify the variable more precisely. In NARK, we list standard single letter variable ‘’bases’’ and an array of prefixes and suffixes to adjust them. Economic variables in model modules should (usually) not use underscores, instead using camel case to the greatest extent possible. For ‘’non-economic’’ variables that are only used temporarily, underscores are permissible. The development team prefers this standard so that users can translate between Python code and LaTeX script with minimal work.
Conventions for naming variables in HARK’s tool modules are significantly closer to more commonly used standards. Variable names should be in all lower case, with underscores between words, e.g. data_to_match. The functions and classes in these modules are more general and almost surely do not have any inherent ‘’economic content’’; they are numerical or algorithmic objects, not variables that might appear in an equation in an article for a (non-computational) economics journal. Variable names in application modules (e.g. the files that execute the cstwMPC estimations) are a mix of the conventions for tool and model files, as appropriate for each variable. That is, variables that are directly related to ‘’economic variables’’ in model modules should follow those conventions, while objects created solely for data manipulation or reporting should use the style of tool modules.
The HARK team wants the toolKit to be as accessible to users as possible; our greatest fear (other than spiders, of course) is that a new user will open up our code, get hopelessly confused trying to read it, and then never look at HARK again. To prevent this tragic outcome, we have tried hard to provide comprehensive, accurate documentation and comments within the code describing our methods. Moreover, HARK uses the Sphinx utility to generate a website with online documentation for all of our tool and model modules, so that users can learn about what’s available in HARK without digging through the source code. When making contributions to HARK, the development team asks users to format their inline documentation to work with Sphinx by following a few simple rules.
The top of every module should begin with a ‘’docstring’’ providing a clear description of the contents of the module. The first sentence should concisely summarize the file, as it may appear in an index or summary of all modules without the remaining sentences. A docstring at the top of a module should be formatted as:
""" Specifies an economic model and provides methods for solving it. More specific description of the key features of the model and variations of it in this module. Maybe some comments about the solution method or limitations of the model. Your bank account routing number. """
The line directly below the declaration of a function, method or class should begin a docstring describing that object. As with modules, the first sentence should concisely summarize the function or class, as it might be included in an index or summary. For functions and methods, the docstring should be formatted as:
def functionName(input1,input2): """ Concise description of the function. More details about what the function does, options or modes available, and maybe mathematical methods used. Credit to a source if you poached their algorithm. Parameters -------------------- input1: type Description of what input1 represents. input2: type Description of what input2 represents. Returns -------------- output_name: type Description of the output(s) of the function. Might have multiple entries. If no output, this is just "None". """
Provide ample comments within a function or method so that a relatively intelligent reader can follow along with your algorithm. Short comments can follow at the end of a line, longer comments (or descriptions of the step or task about to be performed) should precede a block of code on the line(s) above it.
Finally, if you write a new model module, the HARK team asks that you also provide a short mathematical writeup of the model as a PDF. This document does not need to go into great detail about the solution method for the model or the functions and classes included in the module, merely specify the economic model and provide a summary of how it is solved. See ConsumptionSavingModels.pdf for an example of this.
HARK has a test suite that ensures correct
execution on your system. The test suite has to pass before a pull
request can be merged, and tests should be added to cover any
modifications to the code base.
We make use of the pytest and unittests
testing framework, with tests located in the various
pytest, ensure that the library is installed in development mode
$ pip install -e .
Now, run all tests using
$ pytest HARK
Or the tests for a specific submodule
$ pytest HARK/ConsumptionSaving
Or tests from a specific file
$ pytest HARK/ConsumptionSaving/tests/test_ConsAggShockModel.py
HARK uses pre-commit to ensure that all code is well organized and formatted, as well as to ensure the integrity of example notebooks. To install pre-commit, run
$ pip install pre-commit $ pre-commit install
Once you do this, it will run the pre-commit hooks on every commit while only affecting modified files. If you want to run the pre-commit hooks manually on every file, run
$ pre-commit run --all-files
Because this is an optional feature, it does not come installed with
HARK by default. This is to avoid overhead for new users and contributors who might be new to github and software development in general.
If you are having issues with pre-commit, and just want to commit your changes, you can use the
--no-verify flag to bypass the pre-commit hooks.
$ git commit -m "commit message" --no-verify
If you do this, please alert one of the core developers so that we can review your changes to make sure that there are no issues and that your code is formatted correctly.
The following pre-commit hooks are currently configured:
[jupytext] sync, clean up, and execute jupyter notebooks
[black] format code
[pyupgrade] update small python snippets as we drop older versions of python
[blacken-docs] format documentation
[isort] sort imports on .py files
[mirrors-prettier] clean up and format other types of files in codebase
[pre-commit-hooks] other small clean-up/formatting
If you are interested in using pre-commit, please see the pre-commit documentation for more information.
Pull request codes¶
When you submit a pull request to GitHub, GitHub will ask you for a summary. If
your code is not ready to merge, but you want to get feedback, please consider
WIP: experimental optimization or similar for the title of your pull
request. That way we will all know that it’s not yet ready to merge and that
you may be interested in more fundamental comments about design.
When you think the pull request is ready to merge, change the title (using the
Edit button) to remove the
Please report bugs on GitHub.
Developer’s Certificate of Origin 1.1¶
By making a contribution to this project, I certify that:
(a) The contribution was created in whole or in part by me and I have the right to submit it under the open source license indicated in the file; or
(b) The contribution is based upon previous work that, to the best of my knowledge, is covered under an appropriate open source license and I have the right under that license to submit that work with modifications, whether created in whole or in part by me, under the same open source license (unless I am permitted to submit under a different license), as indicated in the file; or
(c) The contribution was provided directly to me by some other person who certified (a), (b) or (c) and I have not modified it.
(d) I understand and agree that this project and the contribution are public and that a record of the contribution (including all personal information I submit with it, including my sign-off) is maintained indefinitely and may be redistributed consistent with this project or the open source license(s) involved.