- Sources
- Run options
- Results
- Notes for developers (!)
Robert Koch institute (RKI):
Case data (RKI-C)
Robert Koch-Institut (2021): SARS-CoV-2 Infektionen in Deutschland, Berlin: Zenodo. DOI:10.5281/zenodo.4681153.
We download the data from github: https://github.com/robert-koch-institut/SARS-CoV-2-Infektionen_in_Deutschland
If the data on github is not available we download the case data from rki from https://npgeo-corona-npgeo-de.hub.arcgis.com/datasets/e408ccf8878541a7ab6f6077a42fd811_0
Vaccination data (RKI-V)
https://github.com/robert-koch-institut/COVID-19-Impfungen_in_Deutschland
Robert Koch-Institut (2021): COVID-19-Impfungen in Deutschland, Berlin: Zenodo. DOI:10.5281/zenodo.5126652
Testing Data (RKI-T)
Hospitalization data (RKI-H)
Population data (P) like "Einwohnerzahl" for Bundesländer and Landkreise:
https://opendata.arcgis.com/datasets/abad92e8eead46a4b0d252ee9438eb53_1.csv
https://www.destatis.de/DE/Themen/Laender-Regionen/Regionales/Gemeindeverzeichnis/Administrativ/04-kreise.xlsx;?__blob=publicationFile
Data from DIVI Intensivregister (DIVI)
Commuter Data from "Bundesagentur fuer Arbeit" (BAA)
Data from "Stastistischen Bundesamt" destatis (DES)
https://www.destatis.de/DE/Themen/Laender-Regionen/Regionales/Gemeindeverzeichnis/_inhalt.html
Data from John Hopkins University (JH)
For informations on installation, dependencies and how to rum the scripts, see epidata README of the above folder.
There are several optional run options
optional arguments working for all are:
optional arguments working for some are:
| -p, --make-plot | Plots the data. |
| -ed, --end-date | Changes date for which data collection is stopped |
| -sd, --start-date | Changes date for which data collection is started |
| -i, --impute-dates | Returns dataframes with all dates instead of only dates where new cases have been reported. Note that this option will have a negative impact on performance as well as on the storage space needed. |
| -m N, --moving-average N | The central N days moving average is computed for the data. Note that the --impute_dates option will be implicitly turned on, as computing the moving average requires all dates to be available. |
| -sb, --split-berlin | Berlin data is split into different counties, instead of having only one county for Berlin. [cases] |
| --rep-date | The reporting date will be prefered over possibly given dates of disease onset. [cases] |
| --sanitize-data | Different ways to distribute vaccinations to home locations of vaccinated persons[vaccination] |
| --files | Files to write [case] |
Hint: When using the "--make-plot" option close one figure-window to get the next one.
The data is written either in json or hdf5 format
The number of "infected" persons is exported as cumulative sum such that "infected" also includes already recovered or deceased persons. Note that for Germany, vaccinations were not reported with the home county of the vaccinated persons but with the county of vaccination.
Note for DIVI:
Not every hospital is reporting the number of corona patients in intensive care units (ICU). The number of reporting hospitals differs from day to day and is given in FullData_DIVI.
More detailed information can be found in the documentation of the different functions.
If a new functionality shall be added please stick to the instructions in epidata README of the above folder.
For information about testing, coverage, pylint and tools see also the epidata README of the above folder.