Ideas of algorithms for which Pythran gives good results?

[paugier]

Hi Serge,

It's not an issue but rather a question (yes the mailing list could have been more appropriate, sorry Github is easier). We're working on py-edu-fr, a new project to create and organize Python trainings in French speaking countries. We'd like to propose a training on "advanced Python for science" and to present Pythran.

We already have few ideas of algorithms to use for teaching Python-Numpy optimization but I was wondering if you could have other good ideas, for example for which Pythran gives nice results.

I thought I would find evident things in https://serge-sans-paille.github.io/pythran-stories but it's not so clear.

[serge-sans-paille]

Maybe have a look at https://github.com/serge-sans-paille/numpy-benchmarks ?
[edit: and hi ;-) If you want I can also review the teaching material]