geodacenter / openairq-dashboard Goto Github PK

Error in .local(.Object, ...) : 

Error in .rasterObjectFromFile(x, objecttype = "RasterBrick", ...) : 
  Cannot create a RasterLayer object from this file.

Confirmed broken across machines, and at the shinyapps.io page. Last successful build was, to my knowledge, 271d78a

Add Chicago Health Atlas tab/ replace vuln. community tab
Determine final Vulnerable Community tab specifications and implement

fb9ff8b added multiple models to the neural net tab, and also varies the color palette by model. While this palette-switching allows better visualization within a given model, I feel it detracts from comparisons between models. Perhaps it is best handled by another selector.

Pls follow up with @james to confirm.

After reworking the EPAPoints argument for dashMap, the lacView and chiView functions' EPAPoints arguments no longer work properly.

The AWS server doesn't launch in some conditions, per @Makosak
Tested and works for me on Safari desktop. Received timeout on Firefox desktop, and inconsistent behavior on mobile, sometimes working, sometimes failing. On Firefox desktop, I'm able to access the port, so likely not a permissions issue, but loading fails when the dash is opened. Any further information is welcome.

Also, update content and combine with downloads page

Update PM Model Legend title from NN to PM 2.5 with units.

--> check remaining maps

Data footers appear a week ahead of the time slider, or vice-versa

Found a bit of unintuitive behavior while going over the dash; clearing the selection on the home tab does clear the selected geography, but as far as I can tell, there is no way to clear the variables selected in the plotly element. Intuitively, I expected clear selection to clear both variables and the selected regions.

Same for precipitation!

Per city:

1. Add helper text to define the variable & method for which data was processed and aggregated here. Link to IDW pages of individual variables if appropriate (refactored as overlaying points/IDW per map)
2. Summary statistics/rolling averages
3. Add NAAQS (or other) benchmarks on plotly, other scales

As of commit d4a1466 the neural net tab crashes the app; appears to be related to the sliderProxy function.
The error thrown is Error in dispatch: Invalid map parameter

Work through issues of leaflet to add graduated color symbology on sensor locations.

-> Show sensor locations (layer as a new point layer in Leaflet)
-> Update so that sensor color changes according to PM 2.5 level
-> Ensure that the color palette is stable/fixed across time to make it more comparable

The range a color palette extends to is determined only via the range of the interpolated raster values (palFromLayer) or the range of the observed readings (palFromVectorLayer). This can lead to some funky maps like below where the range of the Pm2.5 raster values for 05/2016 are 5-10, but there are observed points above 10.

Per city:

1. Color schemes hard to distinguish; either add circles to legend for sensors, or choose more contrasting schemes -> make colors easier to distinguish, add dots to legends
2. Link Chicago Health Atlas

Due to a change in the working of the EPApoints argument of the dashMap function, the FAA-based meteorological maps no longer work. FAA-based points data needs to be updated and integrated into the dashboard.

Currently green implies good, but is bad!

This graph has an error referencing tibble sizes right now if viewed locally.

If viewed on the hosted app, it just says there's a generic error

Per city: Desire to have different levels of spatial/temporal aggregations (incl. as vector), and the ability to compare between them; also consideration of different summary stats, e.g. max/mean