NEXT-Net is a python module can simulate non-Markovian epidemics on arbitrary networks. NEXT-Net is based on a pybind11 wrapper around the C++ simulator https://github.com/oist/NEXTNet
Download the latest release and in a new terminal window run:
pip install nextnet-0.2.0.tar.gzIn a new terminal window run:
#Download the repository
git clone [email protected]:oist/NEXTNetPy.git
cd NEXTNetPy
# Download the required dependencies (boost, pybind11)
git submodule update --init --recursive
# install the python package
pip install .Here is an example of how to run a SIR simulation on a Barabasi-Albert random graph using networkx.
import nextnet as nn
import networkx as nx
import matplotlib.pyplot as plt
n = 10**5 # size of the network
m = 1 # number of edges added per node
graph = nx.barabasi_albert_graph(n,m)
# Define the distribution for the infection times (Gamma distributed)
MEAN_INFECTION = 5
VARIANCE_INFECTION = 1
psi = nn.transmission_time_gamma(MEAN_INFECTION,VARIANCE_INFECTION)
# Define the distribution for the recovery times
MEAN_RECOVERY = 14
VARIANCE_RECOVERY= 3
rho = nn.transmission_time_lognormal(MEAN_RECOVERY,VARIANCE_RECOVERY)
# To simulate a SIR epidemic with one initial infected individual
times, infected = nn.simulate(graph,psi,rho,SIR=True,initial_infected=1)
plt.plot(times,infected)
plt.show()
TBA