CSSEGISandData / COVID-19

Novel Coronavirus (COVID-19) Cases, provided by JHU CSSE
https://systems.jhu.edu/research/public-health/ncov/
29.14k stars 18.43k forks source link

Mainland China time_series_19-covid-Confirmed.csv is missing on 3/11/20 #534

Closed kechan closed 4 years ago

mainadwitiya commented 4 years ago

exactly my whole web-app is messed up now

rks125 commented 4 years ago

It's not missing. They just changed the name from "Mainland China" to "China". I don't get why data sets are changing the structure over the past 2 days.. I recommend using Lat/Long for consistency (do a lookup based on Lat/Long)

mainadwitiya commented 4 years ago

Nope, country/Region is still "Mainland China" in time series data. If you go to the last column, you can check that there is no data available for 03/11/2020

FAGundum commented 4 years ago

@mainadwitiya , the 03/11/2020 data for China is available at the very end of the file, where the country name is called China instead of Mainland China

mainadwitiya commented 4 years ago

@FAGundum thanx I just saw it but still the timeseries is not updated for previous dates, it's messed up right now, the admins have notified that they are working on the dataset

kechan commented 4 years ago

The file time_series_19-covid-Confirmed.csv and others seemed updated. It is now just "China" (without mainland). It seems they roll Hong Kong into it, but kept Taiwan as a country with an asterisks. I am not sure if this is the final state of their fix. I should have kept a backup as a sanity check on if the previous dates are correct.

But just to see if someone can sanity if this looks right. I did a semilogy plot of a few countries and attached here. Italy is perilous close to China in terms of Active Cases, is this right?? Screen Shot 2020-03-12 at 3 06 11 AM