Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/benzhaomin/covid-ddpb
A look at the daily death per inhabitants caused by COVID-19 over time.
https://github.com/benzhaomin/covid-ddpb
covid-19 dataset hacktoberfest
Last synced: 28 days ago
JSON representation
A look at the daily death per inhabitants caused by COVID-19 over time.
- Host: GitHub
- URL: https://github.com/benzhaomin/covid-ddpb
- Owner: Benzhaomin
- License: gpl-3.0
- Created: 2020-03-22T14:11:11.000Z (almost 5 years ago)
- Default Branch: master
- Last Pushed: 2021-07-17T07:17:37.000Z (over 3 years ago)
- Last Synced: 2024-11-09T16:48:10.428Z (3 months ago)
- Topics: covid-19, dataset, hacktoberfest
- Language: Python
- Homepage:
- Size: 3.49 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# COVID DDPB
A look at the daily death per capita caused by COVID-19 over time.
Sources:
- [herman6888/Wuhan-2019-nCoV](https://github.com/herman6888/Wuhan-2019-nCoV).
- which itself uses [今日头条/Toutiao](https://i.snssdk.com/forum/home/v1/info/?forum_id=1656784762444839)
Publications:
- [CoronaWhy Dataverse](http://datasets.coronawhy.org/dataset.xhtml?persistentId=doi:10.5072/FK2/RNGBW9)
## Rationale
The basic idea is that confirmed cases numbers can't be relied on, due to:
- wildly varying amount of tests per country and over time (some countries don't test, then test a lot, then just give up on testing)
- high variance in test protocols, methods and reliability
- cooked numbers
The death numbers are closer to the truth, until localities get completely overwhelmed like in Bergamo, Italy according to the mayor [source](https://www.agi.it/cronaca/news/2020-03-18/coronavirus-quanti-morti-davvero-bergamo-7648225/).
Absolute numbers aren't that useful when comparing countries either so we use daily death per 1 billion people or `ddpb`.
Starting at ~30 ddpb (moving average over 3 days), we can say the epidemic is really underway, call it Day 0. We can then align countries and compare them over time.
## dataset.json
[dataset.json](dataset.json) is a map of `string: list[int]`. One datapoint per country per day, since our Epoch (set to `2020-01-23`).
```json
{
"Country A": [0, 1, 3, 9],
"Country B": [0, 0, 0, 2]
}
```
## dataset.csv
[dataset.csv](dataset.csv) is the same as [dataset.json](dataset.json) with headers on top.
```csv
Country, 2020-01-23, 2020-01-24, 2020-01-25, 2020-01-26, ...
Italy, 0, 1, 3, 9
China, 0, 0, 0, 2
```
## Analysis
See [analysis.md](analysis.md) for an up-to-date summary of the state of the pandemic based on the DDPB metric.
## Dev
Requirements
- jq
- flake8
Setup
```bash
python3 -m venv venv
source venv/bin/activate
pip3 install flake8
```
Makefile
- `make` to update the dataset
- `make test` to test build.py