This is my Bootcamp Mid Project where I created a dashboard about COVID-19 in Germany and Venezuela. It is my first real project since I jumped into data analytics field. You are going to find how I did it and the goals that was setted at then end of it. I hope you like it 🤟
- Create Api in FastApi
- Create Dashboard in streamlit
- Database in MongoDB or PostgreSWL
- Use of geospatial data and geoqueries in MongoDB or Postgres (Using PostGIS)
- Have the database in the Cloud (There are free services in MongoDB Atlas, Heroku Postgres, dentre others)
- Generate pdf report of the data visible in Streamlit, downloadable via button.
- A multi-page dashboard in Streamlit.
- Have the dashboard send you the pdf report by e-mail
- To be able to upload new data to the database via API (username and password as request headers)
- To be able to update the database via Streamlit (with username and password, in a separate page. The dashboard must make the previous request that adds data via API)
- Create Docker container and deploy the services in the cloud (Heroku. The two services must be uploaded separately)
For this project I used this dataset from Kaggle Time Series Data Covid-19 Global. You can find the csv files within data folder: confirmed_global.csv, deaths_global.csv and recovered_global.csv.
I did some data cleaning to get only cases form Germany and I dropped the Province/State column from the dataset because it was empty, at least for this country. Then, I created three new files as the original dataset but named different: germany_confirmed_cases.csv, germany_confirmed_deaths.csv and germany_confirmed_recovered.csv.
If you want yo explore how I did the data cleaning, check the file within data folder or clicking here.
The datasets cleaned was uploaded to Mongo DB as a CSV file for being processed through the API.
I created an API with its routers to extract data when it is neccesary.
The dashboard was created using Streamlit interface.
-
Clone repository
-
Run API
uvicorn main:app -
Run Streamlit
streamlit run main.py
I'd like to thanks to my teachers Santino Lede and Daniel Alvarado for support me in each step of this project. Also, I want to thanks Nicolas Manduley repository, because it helped me to chart my data.
I am a Junior Data Engineer passionate about analyzing data, creating dashboards and data driven cultures.
