Coronavirus India Forecast: SIR model + ML regressions with python

In the context of the global COVID-19 pandemic, Kaggle has launched several challenges in order to provide useful insights that may answer some open scientific questions about the virus. This is the case of the COVID-19 Global Forecasting, in which participants are encouraged to fit worldwide data in order to predict the pandemic evolution, hopefully helping to determine which factors impact the transmission behavior of COVID-19.

Repo: https://github.com/bsivanantham/covid

Code & Visualization: https://www.kaggle.com/balavivek107/covid-india-forecast-sir-model-ml-regressions

Disclaimer: Here I will give the visualization for a quick read and some observation from my side, which just my perspective. I am not an expert.

Exploratory data analysis (EDA)

First, let’s take a look at the data structure:

display(train.head(5))
display(train.describe())
print("Number of Country_Region: ", train['Country_Region'].nunique())
print("Dates go from day", max(train['Date']), "to day", min(train['Date']), ", a total of", train['Date'].nunique(), "days")
print("Countries with Province/State informed: ", train[train['Province_State'].isna()==False]['Country_Region'].unique())