Data repeated on 3/11/20 and 3/12/20

CSSEGISandData / COVID-19

Novel Coronavirus (COVID-19) Cases, provided by JHU CSSE

https://systems.jhu.edu/research/public-health/ncov/

29.11k stars 18.39k forks source link

Data repeated on 3/11/20 and 3/12/20 #804

Open selvaggi opened 4 years ago

selvaggi commented 4 years ago

for example deaths in italy are the same ofr these two dates.

JimBudde commented 4 years ago

I agree, something wonky occurred in the 3/12/2020 data set. Italy's number clearly duplicated from previous day.

Located the version when the duplicate Italy data was introduced; between the these two commits: 3/12/2020 6:01:02pm and 3/12/2020 6:11.07pm commit.

2020-03-15_15-16-35

jescuderoma commented 4 years ago

This problem affects the data from 3-12-2020 of, at leat, Italy, Spain, France and Germany.

showdownhero commented 4 years ago

All the data in the Australian states was impacted by this too

nagyben commented 4 years ago

Same issue observed with UK

llippens commented 4 years ago

Duplicate of #619 (and others)

The author addresses the issue in #650

The issue is known but corrections have not been implemented yet. Hopefully JHU CSSE will commit corrections in the near future.

Hikmer commented 4 years ago

Somewhat unacceptable that this data is so off the mark. People are in panic based on bad data.

sgt101 commented 4 years ago

This is not fixed for me yet

delcacho commented 4 years ago

It happened again today.

sgt101 commented 4 years ago

This is really worrying me now - I am seeing infographics with projections based on fit to the repeated data points circulating. I think that this could influence decision making and could be fixed easily. Why is this being ignored when I have put a pull request and I am sure others have? This is really poor.