opencog / rocca

Rational OpenCog Controlled Agent (ROCCA). Use OpenCog to control a rational agent in OpenAI Gym and Malmo environments.
GNU Affero General Public License v3.0
32 stars 18 forks source link

CI setup on GitHub Actions, nbdev and a basic test #27

Closed ntoxeg closed 3 years ago

ntoxeg commented 3 years ago

So, this is a CI setup that I proposed in #25.

Maybe I will figure out how to run Gym / MineRL code - like getting to run the workflow as a normal user or some other way. Thus, for now checking if notebooks execute is disabled.

ntoxeg commented 3 years ago

Rebased

ntoxeg commented 3 years ago

@ngeiswei review please.

ntoxeg commented 3 years ago

Since I’m adding formatting changes on top of the CI setup, I’ve included the devcontainer setup as well - it doesn’t change anything for anyone not using it but can be nice for new users (and I could finally stop awkwardly throwing these file around without adding them to the repository).

ntoxeg commented 3 years ago

As for the Makefile, it’s not necessary - it just comes from the nbdev template. For now most of its functionality is not relevant (like generating documentation), so I can remove it if you wish. If black is to be used though you can now use it to conveniently generate code from notebooks and format it at the same time.

As for the formatting, it’s your call @ngeiswei - I can revert it and just fix the missing newline manually (which will have to be done every time code from notebooks is generated until the Issue I posted is resolved in nbdev). One important change now is that everyone will have to format the code with black, so if you don’t like the idea of enforcing that I can drop it.

ngeiswei commented 3 years ago

All seems to function, I'm merging. Thanks again @ntoxeg !