Feedback before Code Review 2

[adrhill]

Hi all,

here's some quick feedback before the second code review session next week. I'm opening this as a single issue to not spam your repository.

Documentation

• Please describe in the README what the repo is about and that it is part of course work at TU Berlin
• Please build a doc page until code review session 2 and include a usage example

Code

• Good job on adding compat entries for all of your dependencies!
• It's good practice to include all of your source files, all of your dependencies and all of your exports in your "main file" src/JML_XAI_Project.jl. An example is given here. You don't need to add more modules inside of your package.
• Your LIME struct could use a docstring
• Docstrings need to be written directly above methods and structs, without a newline (here in L36): https://github.com/e-strauss/JML_XAI_Project/blob/4c4c11cf4e2d3063461bb72393fc6b41b83a3a53/src/Lime-base.jl#L29-L40

[e-strauss]

thanks @adrhill , we will look into that.

I have question regarding the unregistered LARS algorithm package we depend on. I looked into other lasso path packages, but they result in a different Coefficient - matrix, which is not suitable from my perspective. So we would stick to the LARS repo for now. I actually I needed to create my own fork of the repo, fix some function renaming to make julia 1.10 compatible...

[adrhill]

So we would stick to the LARS repo for now.

Sounds good!

I actually I needed to create my own fork of the repo, fix some function renaming to make julia 1.10 compatible...

Bonus points if you make a PR to the original repo with your fixes! ;) You can run CI on the latest Julia release (currently 1.10) by updating these lines in LARS.jl to:

- '1.0'
- '1.6'
- '1' # latest release

However, it looks like a similar PR might already exist: https://github.com/simonster/LARS.jl/pull/10

[e-strauss]

thanks - fixed in #51 & #61