search

CSSEGISandData / COVID-19

Novel Coronavirus (COVID-19) Cases, provided by JHU CSSE
https://systems.jhu.edu/research/public-health/ncov/
29.12k stars 18.4k forks source link

US Recoveries missing for March 18th #1075

Open arik-so opened 4 years ago

arik-so commented 4 years ago

Yesterday's update put them at 106 for March 18th, but with today's (March 19th) update, they are at 108 today, and dropped to 0 yesterday.

Additionally, since yesterday's update, per-state recoveries are incorrect. They were correct up until March 17th.

ryanwoconnor commented 4 years ago

Same as #1082 . We need this fixed please.

balazsboros commented 4 years ago

My suggestion would be to completely remove the data that refers to "County, State, US" for the first few cases in the US. It got out of control around 10 days ago, when the number of cases started growing crazily. It is totally fine to keep track only on the state level. What I do (in Matlab code) is that I collect all those old data with counties under the state. But this should be done once and for all at the source, namely at JHU. This would eliminate a potential point of failure from their system, and at the same time would ease the job of tens of thousands of people in the world. As a positive side effect, they could get rid of commas within a cell, which is not very fair to have in a .csv (comma separated values) file.

ryanwoconnor commented 4 years ago

I think we just need correct summarized data for the latest time series and also for 3/18 and 3/19.

Whichever way that can happen, I’m open to. I just don’t want to have to add additional code to massage data out of the box. That causes error prone code.

Let’s make sure we stay focused on the request. My issue doesn’t relate to County level data, I simply want correct values for Stare level data in the recovered csv file.

Thank you! Ryan