1c3t3a/covid-19-analysis

Analyses and visualizes WHO data of the COVID-19 pandemic.

covid-19-analysis's Issues

Change behavior of reloading the csv file to avoid multiple reloads

Therefore change download_csv_file method in CovidCases.py to return None. All other modules need to be changed as well.

When drawing multiple plots in once the first plot will not be formatted as it should be

The first plot will not show a title and the grid is missing. The second plot is okay.

`
fig, ax = plt.subplots(2, 1, figsize = (12,16))
for axis in ax:
axis.xaxis_date()

mi.plot(ax = ax[0])
plt.yscale('linear')
plt.title('Cummulated cases since 31.12.2019')
plt.grid(True)

mi.plot(ax = ax[1])
plt.yscale('log')
plt.title('Logarithmic cummulated cases since 31.12.2019')
plt.grid(True)

plt.show()`

Implement a RESTful API to get the plots

Adding functions to generate a world map of the data

Add an extrapolation for the number of active cases

Create a changelog

Introduce a cluster algorithm to get an overview of the severity of the pandemic in different countries

Actually we have done some work on this that's yet unpublished. The key part would be to find fitting characteristics to group by (our problem was always that non-normalized data always clustered small countries). We used KMeans cluster algorithm, but feel free to suggest another one.

Adapt changes of #64 to REST API

See #64 regarding the REST API

Vertrauenswürdige Quellen ausmachen

Siehe Daten der WHO

Add unit tests for the REST API

The API is written using FastApi. If you look at their website, they have a special section that describes testing here.

Add a function to get the data of the last N days

Adding support for the death rate in percent of the cases

Cleanup files on GitHub

Über Python Data scraper informieren und in BeautifulSoup einlesen

Reload new data on every REST request

Add a Github action workflow

BUG: some calculation are missing the last element of a range

Add a text field to add a group of GeoIDs

Instead of having a hardcoded list of countries being displayed it would be nice to have a comma separated list of GeoIDs as a input text field in Jupiter. The list should be of any length. The Python file should be also updated to have such a list of countries.

REST AOI sample uses fix JSON file and does not download new files automatically

Add an attribute to show the 7 day average number of daily cases per 100000 citizen

The graph of cases per day should be a bar chart with labels

The documentation can be found here: Matplot documentation

Add action workflow for the class test snippet

Add modifications of #57 to the rest API and provide means of using 7 day lowpass filtered daily cases plots

C# sample should display the URL that it is opening

Adding a button in the notebook to download the JSON file

The file can be found here: ECDC website

Add optional axis labels

Toe ensure to display axis information when showing results of #29

Add extrapolation for the reproduction rate R0

This might be difficult but an extrapolation would be sufficient

Modify interface to be compatible with different datasources

Currently we're using the [ECDC] (https://www.ecdc.europa.eu/en/publications-data/download-todays-data-geographic-distribution-covid-19-cases-worldwid) data. It would be nice to have an interface which adds a level of abstraction to "plug-in" different datasources such as local data sources as listed in the Excel file here.

Add Docker information to the readme, WHO changed population data to 2019 data, WHO changed GeoID for the ship in Yokohama

The Who changed the population numbers from 2018 to 2019 data (June 21st) and also the field name in the JSON file
The Who changed the geoID for the ship in Yokohama from "" to some wired numbers (June 22nd)

Daten der vertrauenswürdigen Quellen automatisch auslesen.

Add a

Change usage of JSON file to CSV file. Rework class to make use of Pandas data frames

REST API fails execution because an import can not be found

Add functions and views of the data starting on the day were the number of cases exceed a given number.

This will align the data of different countries based on exceeding the given number of cases. The x axis is no longer a date as the date might be different for the countries. Instead the x axis now shows the number of days

Projektstruktur aufsetzen

Add means to get the data from mb.cmbt.de

instead of using the REST API through a local server the program could access mb.cmbt.de website alternatively

Apply some PEP8 style guides

Create a Setup for the C# application

Adding functions to automatically download the database

Should be updated at least once a day

Remove scrapers and refactor folder structure

Add a Docker version of the rest API

Rename the 'Cases' attribute to 'DailyCases' in all files using it

Refactor structure

The idea is, to have a Base Interface for all Characteristics that we want to analyze. Concrete characteristics will implement that interface. The plotting part should be done in a separate class so that the notebook ist just using calls to all mentioned classes (adding one layer of abstraction to the project).

Adapt changes of #64 in Jupiter notebook and C#

See #64

Add a C# net client for the REST API

Gescrapte Daten in CSV parsen und translaten

Create method to get a list of available countries (GeoID and Country name)

Add a PyQt based GUI querying the REST API

This could be done in a similar way as the C# client. If you would like to use a different library than Qt for this task, feel free.

Switch from Flask to FastAPI as it's providing better documentation functionality

Review documentation and string parser