Vaex: the perfect DataFrame Library for Python Apps
summary
Vaex claims real interactivity on a single node for datasets larger than RAM — memory‑mapped columns, C++ kernels and lazy evaluation let it stream a 350 GB file and compute aggregations (counts, means, histograms) over hundreds of millions of rows in seconds; it supports GPU acceleration, caching and delay=True batching to reduce passes, exposes a progress context that shows exact disk passes, and shows working dashboards built with Solara, Streamlit and Dash.
AI-written summary · about
chapters (7)
- 0:00Vaex introduction
- 4:28Vaex high performance
- 7:37Vaex demo with GDELT
- 13:37Fast data app with Vaex
- 16:46Optimizing with delay and cache
- 21:16Production use cases
- 23:24Q&A
related talks
38mUncluster Your Data Science Using Vaex
40mBeyond Pandas: The Great Python Dataframe Showdown
33mSkinny Pandas Riding On A Rocket
20mStreaming DataFrames: A New Way to Process Streaming Data
38mExpressive and fast dataframes in Python with polars
16mVisualization and Analysis of Large Scale Datasets with Python
30mMeta-Dashboards: Accelerating Geospatial Web Apps Creation with Voilà
45mInteractive Analysis of (Large) Financial Data Sets