In this project we present a course on epidemy modelling, resulting in an algorithm capable of estimating the parameters of several types of epidemy models. The course is structured as:
- Analytical Models: for now we have the SIR model.
- Stochastic Models: currently with the Read Frost model.
- Data driven analysis: this set of analysis are differentiated by past epidemy analysis (such as the United Kingdom one), and the currently COVID epidemy.
The final content, a model capable of learning from epidemy`s data the defined model structure parameters, can be used by just clonning this repository, and acessing the models folder as:
from models import *
# Size of the population
N = 200000
# The model structure
model_type = ("S", "I", "R")
# Create the model
model = ss.epidemicModel(pop=N, focus=model_type)
# Train the model
model.fit( {"S": s_data, "I": i_data, "R": r_data}, time_vec )
# Predict the outputs
S_pred, I_pred, R_pred = model.predict((S_0, I_0, R_0), time_vec)