Project for the hackaton #CodeVsCovid19
Once we started the project we found troubles finding datasets. There were a a lot of them available in the internet but from different sources. You had to go one by one downloading them, looking at the inner structure of them such as headers, types of data, etc. Therefore, we thought how useful could be to have a central repository of dataset for the investigation of Covid-19, like a bank of data, where you can access freely and download all the available datasets from the internet, see the internal structure without having to download it, design it according to your needs and download it when you think it is perfect for you. Additionally, you can even contribute by uploading your own dataset.
With this project we have learned how to mount docker machines with different configurations as we had one for Angular, another for Nodejs, one more for the database (MongoDB) and the final one to support Python with Flask and Conda. This configuration allowed us to learn how to create a web application using Angular, use a non-relational database like MongoDB, and use Python to clean, transform and visualise a dataset.
In terms of how we build it, as mentioned before, we used 3 docker machines each one providing its own service, that is, an Angular service, a MongoDB service, or a Python with Flask service.
As challenges, we knew very little or nothing about this technologies and this made us not finish the project. We had quite a lot of troubles because the configuration of the dockerfiles were not the appropriate one, or we did not know how to use some technologies and we encountered problems that we had to search on the internet, etc.