def test_determine_nodes_to_migrate():
test_class = dynamic_stochastic_block_model(sizes, probs, migration, exp_par, time_until)
# We overwrite the next migration time variable for every node in the class with 5
[test_class.G.nodes[node].update({"Next Migration Time": 5}) for node in test_class.G.nodes()]
assert test_class.determine_nodes_to_migrate(4) == []
assert test_class.determine_nodes_to_migrate(6) == list(test_class.G.nodes())
def test_update_edges():
# update_edges is called during perform_migration_event
# Network has 3 fully connected subgraphs that are disconnected
test_probs = [[1, 0, 0],[0, 1, 0], [0, 0, 1]]
test_migration = [[0, 1, 0], [0, 0, 1], [1, 0, 0]]
test_class = dynamic_stochastic_block_model(sizes, test_probs, test_migration, exp_par, time_until)
# We overwrite the next migration time variable for every node in the class with 5
[test_class.G.nodes[node].update({"Next Migration Time": 5}) for node in test_class.G.nodes()]
# We perform the migration event for node 201, it will be moving to block 0, and should be connected to nodes 0:100
test_class.perform_migration_event(201)
assert list(test_class.G.neighbors(201)) == list(range(100))
def test_generate_migration_times():
test_class = dynamic_stochastic_block_model(sizes, probs, migration, exp_par, time_until)
results = test_class.generate_migration_times(1)
groups = [i[0] for i in results]
times = [i[1] for i in results]
#test times:
assert times[0] == 0
assert times[-1] > time_until
#test groups:
assert groups[0] == 0
assert all([i < 4 for i in groups])
def test_increment_network():
# Network has 3 fully connected subgraphs that are disconnected
test_probs = [[1, 0, 0],[0, 1, 0], [0, 0, 1]]
test_migration = [[0, 1, 0], [0, 0, 1], [1, 0, 0]]
test_class = dynamic_stochastic_block_model(sizes, test_probs, test_migration, 50, time_until)
# We overwrite the next migration time variable for every node in the class with 5
[test_class.G.nodes[node].update({"Next Migration Time": 5}) for node in test_class.G.nodes()]
# We overwrite the next migration time of 201 to be 0.1
test_class.G.nodes[201].update({"Next Migration Time": 0.1})
# The network is incremented forwards by a time of
test_class.increment_network(0.2)
# We perform the migration event for node 201, it will be moving to block 0, and should be connected to nodes 0:100
# It possible that the node could be migrated twice but really unlikely
assert list(test_class.G.neighbors(201)) == list(range(100))
def test_perform_migration_event():
# We set up the migration probabilities so that we know prob 1 which block the node will migrate to
test_migration = [[0, 1, 0], [0, 0, 1], [1, 0, 0]]
test_class = dynamic_stochastic_block_model(sizes, probs, test_migration, exp_par, time_until)
# We overwrite the next migration time variable for every node in the class with 5
[test_class.G.nodes[node].update({"Next Migration Time": 5}) for node in test_class.G.nodes()]
test_class.perform_migration_event(1)
test_class.perform_migration_event(101)
test_class.perform_migration_event(201)
# Nodes in block 0 move to block 1, block 1 -> block 2, block 2 -> block 0
assert test_class.get_node_current_block(1) == 1
assert test_class.get_node_current_block(101) == 2
assert test_class.get_node_current_block(201) == 0
def test_get_node_next_migration_time():
# Test was written before we supported non-exponential waiting times
# Only test that the vector is of approximately the right form.
test_class = dynamic_stochastic_block_model(sizes, probs, migration, exp_par, time_until)
assert len(test_class.get_next_migration_times()) == 300
assert all([ 0 < time < 100 for time in test_class.get_next_migration_times()])
def test_get_node_memberships():
test_migration = [[1, 0, 0], [0, 1, 0], [0, 0, 1]]
test_class = dynamic_stochastic_block_model(sizes, probs, test_migration, exp_par, time_until)
assert all([membership == 0 for membership in test_class.get_node_memberships(1)])
def test_get_node_current_block():
test_class = dynamic_stochastic_block_model(sizes, probs, migration, exp_par, time_until)
assert test_class.get_node_current_block(1) == 0
assert test_class.get_node_current_block(101) == 1
#Test dynamic_sbm
from NetworkEpidemicSimulation.DynamicNetworks import dynamic_stochastic_block_model
sizes = [100, 100, 100]
probs = [[0.4, 0.001, 0.001],[0.001, 0.4, 0.001], [0.001, 0.001, 0.4]]
migration = [[0, 0.5, 0.5], [0.5, 0, 0.5], [0.5, 0.5, 0]]
exp_par = 10
time_until = 100
test_class = dynamic_stochastic_block_model(sizes, probs, migration, exp_par, time_until)
test_class.generate_migration_times(1)
def test_generate_migration_times():
test_class = dynamic_stochastic_block_model(sizes, probs, migration, exp_par, time_until)
results = test_class.generate_migration_times(1)
groups = [i[0] for i in results]
times = [i[1] for i in results]
#test times:
assert times[0] == 0
assert times[-1] > time_until
#test groups:
assert groups[0] == 0
assert all([i < 4 for i in groups])
def test_get_node_current_block():
test_class = dynamic_stochastic_block_model(sizes, probs, migration, exp_par, time_until)
assert test_class.get_node_current_block(1) == 0