You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The repository contains the code for an autoregressive neural network approach to solve epidemic inference problems on contact newtorks. The patient zero problems, risk assmement or the inference of the infectivity of class of individuals are important examples.
4
4
5
-
Up until now ANNFORE supports the SIR compartimental model on contact networks, more complicated compartimental model can be easly added.
5
+
Up until now annfore supports the SIR compartimental model on contact networks, more complicated compartimental model can be easly added.
6
6
7
-
ANNFORE can compute the probability to each individuals to be susceptible, infected or recovered at a given time from a list of contacts and partial observations.
7
+
annfore can compute the probability to each individuals to be susceptible, infected or recovered at a given time from a list of contacts and partial observations.
8
8
At the same time, it can infer the parameters of the propagation model (like the probability of infection <span>λ</span>).
9
9
10
10
The approach is based on the autoregressive probability apporoximation of the postieror probability of the inference problem. See [here](https://arxiv.org/abs/2111.03383) for more details.
0 commit comments