Scipy 1.13

scipy.github.io

52 points·lorepieri··3 comments

> Interactive examples have been added to the documentation, allowing users to run the examples locally on embedded Jupyterlite notebooks in their browser.

This might sound strange, but to me this is pretty exciting.

I've been looking for ways to include _interactive_ Python scripts on static webpages (such as those made using Jupyter Book [1] or Quarto [1]. Up to now the only way I knew how to run Jupyter notebooks on those pages was through, e.g., Binder or Google Colab or some other separate hosted service. I was at the point of giving up on Python and simply learning Javascript.

But now it seems that Pyodide [3] and Jupyterlite [4] are filling that gap. I'm looking forward to trying this out on some static websites.

[1]: https://jupyterbook.org/en/stable/intro.html

[2]: https://quarto.org/

[3]: https://pyodide.org/en/stable/

[4]: https://jupyterlite.readthedocs.io/en/stable/

I am also very happy to see see Python in the browser get more focus, especially for data analysis and presentation. It has been the one week spot of the PyData stack, and it now seems to be getting better pretty fast. At least good enough for demos, examples, onboarding, basic interactive dataviz. Maybe soon also teaching of data science basics.

> This out-of-band release aims to support NumPy 2.0.0, and is backwards compatible to NumPy 1.22.4.

I am not well versed in the upcoming / in progress 2.0 releases for numpy but it looks fairly large in scope.

Reply on news.ycombinator.com