mishra-lab / covid-gta-surge-planning Goto Github PK

Output similar to static plots currently; user selects a variable and a range of values to run the model on, keeping other variables fixed

Can try doing this as a separate page (https://github.com/open-meta/uiStub)

• clarify catchment area (perhaps move population to hospital params)
• fix "duration subclinical" description
• clarify R0 and Reffective
• clarify frequency of super-spreading events
• check parameters:
  • 'baseline non-severe tested'
  • 'baseline hospitalized tested'
  • 'maximum non-severe tested'
  • 'maximum hospitalized testsed'
  • '# super-spreading events'
• consistent naming (e.g. length -> duration)
• clarify domestic vs. imported

Currently, GH action deployment takes a long time due to installation of dependencies.

Possible solutions/workarounds:

• Try https://help.github.com/en/actions/configuring-and-managing-workflows/caching-dependencies-to-speed-up-workflows

• Can try forcing renv to download from MRAN where possible
  See renv config for details:
  https://github.com/rstudio/renv/blob/0a2e1a2aee2ca37661aa72757eb9675bf3bda285/R/config.R#L95-L97

• Make a docker container with all necessary packages pre-installed