Iodide and Pyodide: Bringing Data Science Computation to the Web Browser
summary
Iodide and Pyodide move real data science into the browser by compiling CPython compiled to WebAssembly and shipping numpy and pandas in the browser so notebooks run without a remote kernel. Live demos show Python↔JavaScript data exchange, matplotlib plots and a Bluetooth EEG + FFT pipeline that mostly works with occasional connection glitches. Performance varies — tight C/NumPy loops run near native while Python-heavy code is slower, up to 10x slower worst case, motivating hybrid use and an optional server for sharing and auth.
AI-written summary · about
chapters (9)
- 0:00Iodide and Pyodide intro
- 3:11Iodide vs Jupyter
- 7:14JavaScript in Iodide
- 8:17Pyodide and WebAssembly
- 13:22Pyodide performance
- 15:27Future Pyodide packages
- 17:35Iodide collaboration server
- 20:53EEG demo with Pyodide
- 23:02Q&A
related talks
41mPyodide: A Python distribution for the browser
45mHow to run Python in the browser
45mThe marvelous world of Python in Browser: a nice surprise
26mPyPy.js: What? How? Why?
39mJupyterLite: run all your code in a web browser using WebAssembly
35mJupyterLite: Jupyter ❤️ WebAssembly ❤️ Python
29mWhat can't WebAssembly do?
19mEmpowering Learners with WebR, Pyodide, and Quarto