Status: Closed (View Workflow)
I think we can update the README file.
Here we use `will` as in the future, but I think we should change it to the present.
>This plugin will be capable of executing code fragments via IPython kernel as currently supported by Jupyter.
Under the Python dependencies, we need to specify the Python version. I can see JDK 8, but no mention of Python version.
Also probably just need a list of dependencies. So users can choose to use `pip`, `poetry`, `conda`, etc. But probably also good to specify the version if we know what version is required.
The section about contributing to the plugin is probably better in CONTRIBUTING.md, in a separate file.
Finally, the instructions start with the Python dependencies, then go on some configuration. Then come back to JDK and more dependencies in "Guideline for setup". We should have a single section for the setup/installations.
I think there is no mention of the Dockerfile in the README. We need to explain how that's used
loghijiaha maybe we should also clarify on the scope of variables when using multiple build steps. Some users that are used to notebooks, may think of build steps as cells, but the variables are not shared right.
(in the second build step, a is not defined)
Yes kinow. This issue we were discussing while merging the IPython builder. It is happening due to Ipython interpreter is closing after executing a build step. For the next build, it opens a new Ipython interpreter. Is there any way to preserve the Ipython instance until a single job finishes? If it is possible I will open an issue for that later.
I remember that now loghijiaha. Let's create a separate issue for that and leave this one only for the README. We can document this as a known limitation.
There you go https://issues.jenkins-ci.org/browse/JENKINS-63073. And we keep the discussion here for the README now
Refer to the link for more information. https://github.com/jenkinsci/machine-learning-plugin/blob/master/CHANGELOG.md
I realized the other installation instructions are for contributing. Probably better keep all that information in the README. And in the contributing file or instructions we can just reference that section.