Hub

Python Playground

The default language of data science

Playgrounds

Sandbox

  • Playgrounds

Languages

  • 🐍Python
  • 🗄️SQL
  • 📊R
  • ⚡JavaScript
  • 🔷TypeScript
  • 🐘PostgreSQL
  • 🔶dbt
  • 📊DAX
  • 🖥️Bash / Shell
  • 📗Spreadsheet Formulas
← All languages
🐍
livedata scienceJupyter Lab (full)

Python Playground

Run Python entirely in your browser with NumPy, Pandas, Matplotlib, SciPy, scikit-learn, statsmodels, SymPy, and more — ideal for cleaning data, building models, and creating charts without installing anything.

In this workspace

What you get when you open Jupyter Lab.

Run All
  • Rich Python stack

    pandas, scikit-learn, matplotlib, and micropip installs

  • Multi-cell notebook

    Add, reorder, and run cells independently or all at once

  • Datasets & samples

    Iris, Titanic, and your own CSV/uploads in a Data explorer

  • AI tutor + cheatsheet

    Inline help and click-to-insert snippets beside the notebook

  • Export & share

    Download notebook formats and share a read-only link

Your Python playgrounds

🚀

No Python playgrounds yet

Opens in the full Jupyter/RStudio notebook workspace with datasets, export, and AI tutor.

Starter template

# Python playground — iris.csv loads automatically from Data
# Need an extra package? Run once per session, then import normally:
# import micropip
# await micropip.install("seaborn")
# import seaborn as sns

import pandas as pd
import matplotlib.pyplot as plt

iris = pd.read_csv("iris.csv")
print(iris.head())
print(iris.groupby("species")[["sepal_length", "petal_length"]].mean())

# Color points by species (map the text labels to numeric codes for matplotlib)
species_codes = iris["species"].astype("category").cat.codes
iris.plot.scatter(
    x="sepal_length",
    y="petal_length",
    c=species_codes,
    colormap="viridis",
    title="Iris clusters",
)
plt.show()

What you'll learn

  • ✓EDA & visualization
  • ✓ML prototyping
  • ✓Statistics homework
  • ✓Data cleaning pipelines
Loading usage…

Runtime requirements

Engine
Pyodide
Type
wasm
Cold start
~25–40 MB (first load)
Execution
Runs fully in your browser

Packages / WASM

  • pyodide@0.23.4
  • numpy
  • pandas
  • matplotlib
  • scipy
  • scikit-learn
  • statsmodels
  • sympy
  • pillow
  • beautifulsoup4
  • networkx
  • micropip

Limitations

  • Pyodide distribution packages auto-load on import; others use micropip once per session
  • micropip supports Pyodide WASM wheels and pure-Python PyPI wheels — not arbitrary C extensions (e.g. TensorFlow)
  • Long-running loops block the worker (30s watchdog)

Libraries

NumPyPandasMatplotlibSciPyscikit-learnstatsmodelsSymPyPillowBeautifulSoupNetworkXmicropip (PyPI install)standard library

Still on the roadmap

  • Progress & challenges

    Micro-challenges per language with XP rewards, similar to catalog tracks.