ICEModel KeyError: 'com'

[jasongong11]

Describe the bug
I run the tutorial code of ICEmodel. And it produces the error code:

  0%|          | 1/200 [00:00<00:00, 853.37it/s]
---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
<ipython-input-17-42574a55ef61> in <module>
     22 
     23 # Simulation execution
---> 24 iterations = model.iteration_bunch(200)

~/anaconda3/lib/python3.7/site-packages/ndlib/models/DiffusionModel.py in iteration_bunch(self, bunch_size, node_status)
    199         system_status = []
    200         for it in tqdm.tqdm(past.builtins.xrange(0, bunch_size)):
--> 201             its = self.iteration(node_status)
    202             system_status.append(its)
    203         return system_status

~/anaconda3/lib/python3.7/site-packages/ndlib/models/epidemics/ICPModel.py in iteration(self, node_status)
     84                         key = (u, v)
     85 
---> 86                         if self.params['nodes']['com'][u] == self.params['nodes']['com'][v]: # within same community
     87                             # Individual specified thresholds
     88                             if 'threshold' in self.params['edges']:

KeyError: 'com'

To Reproduce
The tutorial code I run:

import networkx as nx
import ndlib.models.ModelConfig as mc
import ndlib.models.epidemics as ep

g = nx.erdos_renyi_graph(1000, 0.1)

# Model selection
model = ep.ICPModel(g)

# Model Configuration
config = mc.Configuration()
config.add_model_parameter('fraction_infected', 0.1)
config.add_model_parameter('permeability', 0.3)


# Setting the edge parameters
threshold = 0.1
for e in g.edges():
    config.add_edge_configuration("threshold", e, threshold)

model.set_initial_status(config)

# Simulation execution
iterations = model.iteration_bunch(200)

• NDlib version
• Operating System
• Python version
• Version(s) of NDlib required libraries

Expected behavior
I guess this issue is probably the community information is not inputted into the model.params['node']

The tutorial code does not run any community detection analysis. And the model.params['node'] gives an empty dict.

Additional context

Name: ndlib
Version: 5.0.2
Summary: Network Diffusion Library
Home-page: https://github.com/GiulioRossetti/ndlib
Author: Giulio Rossetti
Author-email: [email protected]
License: BSD-Clause-2
Location: /home/jasongong/anaconda3/lib/python3.7/site-packages
Requires: bokeh, dynetx, future, netdispatch, networkx, numpy, python-igraph, scipy
Required-by: 
Note: you may need to restart the kernel to use updated packages.

[JKORELC]

Hi @jasongong11,

Hopefully you found the answer by now. For anyone else that arrives here for the same reason, there is a simple solution.

   if self.params['nodes']['com'][u] == self.params['nodes']['com'][v]: # within same community
       # Individual specified thresholds
       if 'threshold' in self.params['edges']:
           ...

ICPModel Line #86

The documented example leaves out this crucial detail.

    for n in g.nodes():
        config.add_node_configuration("com", n, community)

Where community is predefined by the community detection algorithm you apply separately.