An 'online poster' for SciPy 2020. More about me on my homepage.

How do you take the experience of a poster session online? I decided to keep a few core elements:

• A traditional poster (as image and PDF)
• Sharing some exciting new developments
• Connections to other topics at the conference

You can read more about Hypothesis on its website, and find the docs here (including Numpy and Pandas support). And of course, you should cite Hypothesis if you use it!

In Praise of Property-Based Testing is the best general introduction I know, while my talk Sufficiently Advanced Testing covers dives into the details of how it works.

I also wrote a paper on property-based testing for science and data code for SciPy 2020, so if you're wondering how or why to get started you can check that out 😉

At the Sprints

Want to get involved at the sprints? You could pick a Hypothesis issue, or help write tests for one of the many scientific projects which use Hypothesis, including:

• Numpy - for example #15189 or #15963
• Pandas - for example #22280
• Xarray - #1846 has a whole list of property-based tests to write
• Astropy is open to Hypothesis tests - for example #9532 or PINT #549.
• replicating metamorphic tests for tools like scikit-learn (see paper for details)
• or your project here?

I believe that working in the open makes both software and science better - and that most scientific software is severely undertested. I am therefore happy to review pull requests or existing tests which apply Hypothesis to open source science projects, or discuss how you might do so. Ping @Zac-HD on GitHub or send me an email!

CC-BY-NC-ND or ask-me-for-a license // back to homepage // forward to more talks