Skip to content

Conversation

ColCarroll
Copy link
Member

This is tweaking #1473 -- test suite works inside the container, and running . ./scripts/start_container.sh works on my machine.

It is ready to merge -- it does not change existing functionality (or add new features -- it is only for ease of development), but I can start a new 3.1 branch with this if we want to be strict about only bug fixes going into 3.0 @fonnesbeck.

@springcoil
Copy link
Contributor

I think this is fine for 3.0

@AustinRochford
Copy link
Member

Thanks for figuring this out @ColCarroll!

CONTRIBUTING.md Outdated

We have provided a Dockerfile which helps for isolating build problems, and local development.
Install [Docker](https://www.docker.com/) for your operating system, clone this repo, then
then run `./scripts/start_container.sh`. This should start a local docker container called `pymc3`,
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

typo then then

@twiecki
Copy link
Member

twiecki commented Nov 14, 2016

Since this is just adding new code I think we should merge this into master after the typo is fixed.

@twiecki twiecki merged commit 4f54823 into pymc-devs:master Nov 14, 2016
@twiecki
Copy link
Member

twiecki commented Nov 15, 2016

I added an auto-build to docker hub: https://hub.docker.com/r/pymc/pymc3/ (let me know if I should add you to the docker org).

So you don't have to build the image anymore, you can just do: docker pull pymc/pymc3

@twiecki
Copy link
Member

twiecki commented Nov 15, 2016

docker run -d -p 8888:8888 pymc/pymc3

Brings up the jupyter notebook, but you can't import pymc3. I now realize that this docker image isn't meant to be used by a user, although we could easily extend/change it to do that.

@ColCarroll
Copy link
Member Author

@twiecki interesting idea. what about building pymc in docker properly (i.e., pip installing it), and having a standalone pymc3-docker project? Then the start-container.sh file could still mount the local pymc and edit the $PYTHONPATH.

My worry is that since it is pretty easy to install, the docker image wouldn't be used, so there isn't much upside.

@twiecki
Copy link
Member

twiecki commented Nov 15, 2016

Yeah, good point. Since we have it working on conda and linux is easy to begin with I agree with your assessment. In any case, we can use the dockerhub for easier debugging of travis problems now.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants