mathLab / PINA

Physics-Informed Neural networks for Advanced modeling
https://mathlab.github.io/PINA/
MIT License
355 stars 62 forks source link

Adding Comments to the Examples #103

Closed yorkiva closed 1 year ago

yorkiva commented 1 year ago

Is your feature request related to a problem? Please describe. The package contains a good set of examples. But their comprehensibility is not up to the mark, especially for someone who is just starting with the package and not yet familiar with the different modules and their utilities.

Describe the solution you'd like In the examples, comments should be added to make it clear what problem is being solved, what are the domains, constraints etc. Additional comments should be added to clearly illustrate what different parts of the code are accomplishing.

Describe alternatives you've considered

Additional context https://github.com/openjournals/joss-reviews/issues/5352

yorkiva commented 1 year ago

In fact, I think it would be superb to convert some of the examples into notebooks. Some of these examples seem small, even possibly trainable on CPUs. It would be good to have notebooks that clearly show the logging during the progress, the kind of outputs it generates, and how to do inferencing and make plots.

dario-coscia commented 1 year ago

👋🏻 @yorkiva Thanks for the feedback! Indeed you are right, we will add comments to the examples. Some of the interesting problems (e.g Stokes) are a bit too long to train on a notebook, but we have a list of possible tutorials here in .py, .html, .ipynb formats. These tutorials comprehend different types of problems (e.g. time dependent or spatial problems). We will add a tutorial for parametric dependent problems, maybe following the Poisson example in example folder which is trained by DeepONet.

dario-coscia commented 1 year ago

👋🏻 @yorkiva . I updated the example folder (see https://github.com/mathLab/PINA/pull/105), so now every problem in example/problem is documented. Few of the problems are not available at the moment.

We will soon release a beta version of PINA where some high level features will change (rename some classes, and add more flexibility for the user by enabling pytorch lightining). Since updating the tutorial requires a big effort (for documentation compatibility, etc.), and due to the soon release of the beta verison, we will not make a tutorial on Stokes (but you can see the working example in example folder`).

Anyway, thank you for the feedback, we will also take into account when working on the beta version #85 .

danielskatz commented 1 year ago

@yorkiva - is this ok to close now?

yorkiva commented 1 year ago

Thanks for the response. This can be closed now.