When is COVID-19 going to end? News is telling a fast-growing number and predicting many different forecasts. I think I can use what we just learned to study for myself. The case number counts data is quite easily available, and the logistic model is a pretty good starting point. The model needs more free parameters to study the different growing trends around the world, and I know Dr. Gong can advise me on this.
I read in the articles that machine learning can be helpful in fighting COVID-19, but I don’t exactly know how to do it yet. There are materials about machine learning techs, and the youtube videos and bilibili videos also look good in teaching myself some ML skills. Good thing it is coded in python.
The total fitting cases in divided fitting part compared with the original cases showed that using the normal logistic model, there are still some errors between fitting of model and the true value. The inset shows the fitting in Belgium had a large tail error because of a second outbreak.