How to build a (small) homemade machine learning model to predict Covid indicators for France

At the beginning…

Simple Data Science stack
Simple Data Science stack
Fig. 1 — Simple Data Science stack

EDA Time ! (Exploratory Data Analysis)

Fig. 2 — Raw data of covid-19 new cases and tests evolution in France
Fig. 3 — Zoom on few weeks
Fig 4. — Applying a 7-days moving average
Fig. 5 — French infection dynamics (7-days moving average)
Fig. 6 — Cumulative french infection dynamics

Machine Learning part

Fig. 7 — The fail of ARIMA methods…
Fig. 8 — Time features building example
Fig. 9 — Train and test periods
Fig. 10 — Models performances
Fig. 11 — Prediction values per model
Fig. 12 — Cumulated Sum of Prediction values per model

…because in the end, I built an entire workflow

Fig. 13 — Tech stack used for whole workflow
Fig. 14 — Airflow ML workflow
Fig. 15 — Data splits for features selection
Fig. 16 — the two pages of the Data Studio dashboard

Final words

Freelance consultant. Data Scientist, AI Architect, Computer science PhD.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store