Next Steps

[ehmatthes]

• Update data
  • Develop a routine for regularly ingesting new data and updating all plots and summaries.
• Generate a static image for the most recent 48-hour period, with critical historical periods superimposed.
  • This would help devise a real-time visualization format.
• Generate static images along with Plotly html files. Plotly html files are giant bodies of JavaScript, and can't always be viewed easily.
  • See kaleido:
    • https://medium.com/plotly/introducing-kaleido-b03c4b7b1d81
    • https://plotly.github.io/plotly.py-docs/generated/plotly.io.write_image.html
• Clean up the code.
  • Refactor, clarify overall logical flow.
  • Should have couple top-level files: process_hx_data.py, process_current_data.py. Rest should probably be in a utils directory.

[ehmatthes]

Consider making an interactive data explorer for this historical analysis. Consider PlotJuggler, which would probably be faster than some browser-based approaches.