Wednesday November 16th 2022: Major advances in HoloViz for Pangeo: GUI explorer to build plots, easy interactive pipelines, and publishing in-browser apps

Announcement
For our last Pangeo showcase before the break, we are pleased to have James Bednar from Anaconda to talk about the new release of the HoloViz libraries and all it’s improvements and what the future may bring.

Meeting Logistics
Title: Major advances in HoloViz for Pangeo: GUI explorer to build plots, easy interactive pipelines, and publishing in-browser apps
Invited Speaker: James A. Bednar at Anaconda (github: jbednar | linkedin | twitter: @JamesABednar | ORCiD )
When: Wednesday November 16th 12PM EDT
Where: Launch Meeting - Zoom
Abstract:
The HoloViz set of open-source Python libraries focus on making common tasks in data exploration, analysis, and visualization easier, faster, cloud-friendly, and more scalable. New releases in October 2022 bring major new capabilities relevant to existing and potential HoloViz users in Pangeo: (1) Panel apps can now be exported to standalone webassembly that runs entirely in the browser, so that you can share apps using simple file-sharing sites like github.io without needing any Python server. (2) Making those apps is now even simpler, using the more-powerful .interactive support of hvPlot – just put widgets into your existing Pandas or xarray pipelines wherever there is a parameter you want to vary, to get a nearly instant Panel app for whatever you are working on right now. (3) Maybe you don’t even need to code anything, because hvPlot now ships with an explorer that you can invoke in a Jupyter cell or as a standalone app that lets you explore your Pandas dataframe (or soon, Xarray Dataset or DataArray) flexibly, outputting the code to invoke that visualization for you to save for later. (4) If you alternate between exploring data, well supported using a Bokeh interactive plot, and publishing results, well supported with a Matplotlib plot, hvPlot now lets you choose to create either Matplotlib or Bokeh plots (or Plotly!) from the same .plot() specification, so that you don’t have to switch tools when you go to publish and you can more easily integrate hvPlot output with hand-constructed Matplotlib plots. All together, these advances make it much easier to explore data, create plots, create apps, and share apps, so that Pangeans can focus on science and saving the planet!
Relevant material:

Agenda:

  • 5-15 minutes - Community showcase
  • 5-15 minutes - Q&A / Community check-in
  • 20-35 minutes - Agenda and Open discussion
4 Likes