Python frontend for visualization and analysis of COVID19 data in Spain
COnVIDa (https://convida.inf.um.es) is a tool developed by the Cybersecurity and Data Science Laboratory at the University of Murcia (Spain) that allows to easily gather data related to the COVID19 pandemic form different data sources, in the context of Spain.
In this repository, you will find everything you need for the analysis and visualization of the data collected by the backend (COnVIDa-lib). This web interface allows you to select intuitively the data you want to analyze and compare it with the rest of the selected data, as well as the possibility of downloading this data in different formats for your own analysis.
We developed the web interface using Flask + Dash, which is enough to run the web application locally.
If your intention is to take this web application to a production environment, it will be necessary to use a Web Server Gateway Interface (WSGI).
On our page convida.inf.um.es we use:
With WSGI we can redirect the user requests to the script in Python, in our case main.py.
For a production environment it is advisable to set the global variable DEBUG = False in main.py.
NOTE: There are other valid alternatives for the deployment of the service.
COnVIDa frontend is composed of two main elements:
main.py: Core implementation of the frontend functionality.COnVIDa-lib: Contains the Python library developed to collect data from the different data sources, COnVIDa-lib
- Python version 3.7 or more, pyenv recommended
- pip3
- Clone the repo
git clone --recurse-submodules https://github.com/CyberDataLab/COnVIDa.gitWith --recurse-submodules you will also get the COnVIDa-lib associated to the frontend.
If you wish, you can choose to download only the frontend
git clone https://github.com/CyberDataLab/COnVIDa.gitand create your own library or adapt ours to new data sources of your interest.
- Change to COnVIDa directory
- (Optional) Create your virtual environment, and activate it (you can also use
condato create the virtual environment)
python -m venv env
source env/bin/activate # Linux/Mac
env/Scripts/activate # Windows- Install required packages
pip3 install -r requirements.txtTo run an app locally:
- Run main.py
python main.py- Open http://127.0.0.1:8899 or http://localhost:8899 in your browser
To change the web service port, change the global variable PORT in main.py
-
https://convida.inf.um.es/help (English)
-
https://convida.inf.um.es/ayuda (Spanish)
The main.py file is the core of the website's functionality. We will find all the functions and parameters needed to modify any aspect of the website. Some considerations should be taken:
-
DEBUGvariable should be changed according to whether we are in a development or production environment. -
The following code included in the main.py allows you to obtain all the data from the COnVIDa library and load it into memory to later display it on the web.
def daily_update():
print("Starting Thread...")
while True:
time.sleep(86400) # 1 day
nAttemps = 0 # 3 attemps
try:
while ((not convida_server.daily_update()) and (nAttemps < 3)):
nAttemps += 1
time.sleep(240)
if (nAttemps < 3):
f = open(os.path.join(FOLDER_LOGS, 'update.txt'), "w")
last_update = dt.now().strftime('%d-%m-%Y, %H:%M:%S (GMT+1)')
f.write("{}".format(last_update))
f.close()
except:
print("Thread Critical Error")We have used a thread that loads the new data into memory once a day with a maximum of three attempts. Once the data have been loaded, the service will be automatically restarted to finish loading the new data.
-
Each one of the defined functions are commented in the code, where a small summary is given about the purpose of such function, as well as its input and output parameters.
-
We have used a Dash functionality called callback for user interaction with the different options on the web. This allows us to declare which is the input parameter (the one that triggers this function) and what is going to be generated as an output.
An example of the use of callbacks would be
@dash_app.callback(
Output("selected_regions", "value"),
[Input("select_all_regions_button", "n_clicks")],
)
def select_all_regions(n_clicks):
if n_clicks > 0:
return [dropdown_option.get("label") for dropdown_option in regions_options]
return []In this case when we click on the button with id="select_all_regions_button", the resulting output will be to add all the region names to the appropriate dropdown.
It contains the style files of the web page, as well as the images, the static HTML pages and the javascript necessary for the correct operation of the web.
- The file convida_dict.json contains the Spanish-English translation for each of the labels on the website. This dictionary allows easily including further languages in the future.
The tmp folder at run time will contain the files, in different formats, with the data you have selected. These files must be deleted periodically in order not to exceed the storage of the server. To this end, we launch a thread that checks and deletes the contents of this folder.
The data on our website are updated daily from each of the sources we have. In each of these updates, we record if it has been done correctly, along with the date and time at which it was done. This is used by our service to know what data can be offered on our website, as well as for the administrators in case there is any error.
Distributed under the MIT License. See LICENSE for more information.
CyberDataLab - @cyberdatalab
Contact us through [email protected]
Entire COnVIDa project: https://github.com/CyberDataLab/COnVIDa-dev