Working interactively with large and remote datasets in...
summary
Three JupyterLab extensions aim to let the UI handle big remote data rather than forcing full eager downloads: @jupyterlab/hdf5 adds a server extension that lazily fetches chunks via h5py and a phosphor/datagrid frontend so terabyte files and a 100 million‑row dataset open interactively; jupyterlab-remote-data experiments with an alternative contents handler; jupyterlab-data-explorer offers pluggable loaders and grid/tree widgets and is the leading candidate to replace the file browser with hooks to inject import snippets into notebooks.
AI-written summary · about
chapters (7)
- 0:00JupyterLab introduction
- 3:13Remote dataset challenges
- 5:37HDF5 integration
- 8:05HDF5 limitations
- 9:07Unified data integration vision
- 10:35Multi-file browser UI
- 11:53Q&A: JupyterLab vs Classic
related talks
29mJupyterLab+Real Time Collaboration
40mJupyterLab: The Evolution of the Jupyter Notebook
40mJupyterLab: Building Blocks for Interactive Computing
37mJupyterLab
28mBuild a data studio in your notebook with jupyter-fs
27mBuilding web-based engineering applications with JupyterLab components
01h 00mInteractive Computing with Jupyter: Past, Present, and Future
49mExperimenting with Large Data Volumes Using Notebooks