Hacker News new | ask | show | jobs
Covidify – corona virus report generator (Python) (github.com)
44 points by AaronWard 2293 days ago
7 comments

This is also a good chart: https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594...

It shows global trend lines, and per-country + per-region data.

That said, nice job to OP.

Thanks very much :))
For weeks I've been hoping to find a way to quickly compare log-plots of numbers infected (optionally normalized to population size) for different countries. IMO the best way to look at world-wide spread, but doesn't seem to be available anywhere. Would be a great addition to this.
I will look into it, if you want could you add it to the issues what exactly you suggest and ill add it to the kanban board :)
Looks cool but what does it actually produce? Graphs? In what format? Images?

The readme could be a bit more clear in what it actually does, but it looks super interesting!

It creates a time series dataset and a summary stats excel report that shows the daily counts, averages, moving averages, deltas, and more (see this excel sheet https://github.com/AaronWard/covid-19-analysis/blob/master/r...)

also, it generates visualizations for 4 attributes - confirmed cases - currently infected - recoveries - deaths

Thanks very much for the feedback :)

Here's another chart I built: https://observablehq.com/@neelance/corvid-19-trends (Yes, typo in the url, can't change it...)
Am I the only one who can't get this damn thing to run on macOS 10.14, Debian testing, or SL7-rolling? Despite pandas being installed I still get 'ImportError: No module named pandas'

¯\_(ツ)_/¯ oh well

Are you using a conda environment?
No, pip everywhere
Check where you have pandas and covidify installed and make sure they are in the same site-packages.

ps: pip is just a package manager, you should really consider using conda for creating environments (it will make your life a lot easier)

Why not just make a website?
Well it can probably be deployed on one. Its a talent showcase thats interesting to others.
It preprocesses the datasource into a time series dataset. So i guess this would be more useful for data science people and ml engineers rather than the average joe
Would be neat if this could also show the same data for the flu.
Do you know of any datasources to integrate flu statistics?
Not sure where the raw data is but there seems to be various things here: https://www.cdc.gov/flu/weekly/index.htm