Celebrating 100 Commits on epispot/epispot
05 Feb 2021
Written By quantum9innovation
100 commits marks a landmark in the epispot development process. Here’s how we got there:
Epispot began as a pile of code for random epidemiological models dumped into a Github repo. Back in these prehistoric times, the epispot organization didn’t even exist!
Epispot’s initial commit added a sir-model.py program that implemented a very basic SIR model
inspired by Henry Ferose’s article on Medium that went on
create the Model class that is so crucial to epispot today. Since then, epispot has come a long way. In celebration of its
hundreth commit which passed epispot v2.0.0, this blog post examines the most important commits
that have brought epispot to where it is today.
Commit #56: The Release of R Naught
After its rough beginnings, the epispot repo was finally beginning to take shape. Code was published to PyPi in package-ready form which allowed users to model COVID-19 anywhere at anytime with the simple command
pip install epispot
This commit sealed the first production release of epispot, R Naught, which features full support for most of the package–even today. R Naught is considered by epispot historians to be the start of the current package as we know it today. The release did not yet have documentation, but the path for epispot was clearer than ever.
Commit #79: Epispot Gets Some Norms
As epispot became more user-friendly and documentation and tutorials materialized, the developers of epispot realized that norms would be crucial to ensure each release delivered users the quality that was expected. One of the great advancements that came from the so-called “Golden Era” of epispot was the invention of the matrix, which looked something like this:
| Packaging | Docs | Features |
|---|---|---|
| ready ✔️ | docstrings ✔️ | compartments ✔️ |
| pip ✔️ | docs ❌ | spatial models ❌ |
| conda ❌ | tutorials 🟡 | interactive plots ❌ |
Every time a new release was published, the status of one of these goals would be changed. While seemingly simple, this was revolutionary. Notice that each column has an equal
number of rows–that’s not a coincidence, it’s a design choice. Epispot engineers would work on each part of the package equally, guaranteeing that epispot always improved
where it was weakest.
Armed with a new tool, epispot engineers were proud to announce the unveiling of epispot v1.0.0, or as it is more commonly known: Herd Immunity. This release changed everything.
Commit 100: The Modern Era
With commit #100, epispot continues its growth, adding new features everyday (and there’s a whole new organization out there!) that change the way epidemiological models are compiled, combating COVID-19 with data science. As epispot begins to expand into new areas of research (Discord Bots?, Project Management?), you can know for certain that epispot will always continue this journey, finding innovative solutions and adding brilliant new features to advance the future of epidemiology.