Skip to content

Latest commit

 

History

History
38 lines (30 loc) · 969 Bytes

altair.md

File metadata and controls

38 lines (30 loc) · 969 Bytes

Altair

A delarative statistical visualization library for Python.

https://altair-viz.github.io/index.html

  • Brian Granger
  • Jake VanderPlas

Why?

  • Lots of visualization options in Python
  • Matplotlib
  • Seaborn
  • Bokeh
  • Plotly
  • BQPlot
  • etc.
  • Very powerful
  • And yet, still lots of pain for users
    • Categorical, datetime datatypes
    • Inconsistent, leaky abstractions and visualization grammar
  • Meanwhile, JavaScript-based visualization advancing at light speed

What?

  • Not a new rendering library
  • A Python API that emits valid VegaLite JSON
  • Rendered in the Jupyter Notebook and JupyterLab by the VegaLite JavaScript library
  • Statistical visualization
    • Data is a DataFrame/Table
    • In the tidy format
    • Mapped to visual properties using groupby operations
  • Most the Python API is autogenerated from the JSON spec