Jupyterlite is a Python/ Jupyterlab distribution that runs entirely in the browser. The various Python packages have been ported to WebAssembly (WASM), a binary format that is executable in browsers. For optimized libraries (numpy, pandas, etc.), performance is equivalent to native performance. For pure Python (loops, etc.), execution is 3 to 10 times slower. What this means, however, is that you can prepare Python programs and Jupyter notebooks that are fully executable in a browser, with zero installation! And it works independently of the system, even on tablets, phones, etc.
For example, you can find here: https://
11.1Deploy to Jupyterlite - 🧑🏫🏫¶
Preparation and deployment under Jupyterlite are not exactly immediate (there are documents to follow and adaptations to make), so I suggest a shortcut, which is as follows:
Download the archive of a deployed version, available here
https://drive .google .com /file /d /14BnzVmPO6I8uOMEmNC6BTMsjXL3RLrQL /view ?usp = sharing
and unzip it on your disk,Place the files you want to make available in
_output/filesRun the program
update_jupyterlite_contents.pyavailable in theextrasdirectory by typingpython update_jupyterlite_contents.py. Install dependencies if necessary.
👉🏼 This will update the lists and characteristics of the files in the distribution
init) Upload all of the contents of
_outputto your web account, typically (or for instance...) tologin/public_html, in a directory such asMyJlite, and you will obtain a distribution that is accessible and executable athttps://www.esiee.fr/~login/MyJlite. If you don’t know how to do this, ask your nearest local IT guru.update) If you only need to perform an update, simply copy the contents of
_output/filesand_output/apitologin/public_html/MyJlite/filesandlogin/public_html/MyJlite/api. For subsequent updates, add or modify files, you only need to perform steps 2, 3, and 5.maj); this will save you from having to upload the 70 MB in step 4.init).