yawning_titan.networks.network_creator#

Network Creator module enables the creation of pre-defined types of networks.

Networks that can be created are:

  • The standard 18-node network from the Ridley 2017 paper.

  • The DCBO network used in the RL Baseline module.

  • A mesh network.

  • A star network.

  • A ring network.

  • A P2P network.

  • A randomly generated binominal network.

  • A custom network using user-input options.

Functions

check_if_nearby

Check if a randomly generated point is close to points already generated.

create_mesh

Create a mesh node environment.

create_p2p

Create a two group network.

create_ring

Create a ring network.

create_star

Create a star node environment.

custom_network

Create custom network through user interaction.

dcbo_base_network

Creates the same network used to generated DCBO data.

generate_node_positions

Generate a random position for each node and saves it as a dictionary.

get_18_node_network_mesh

The standard 18 node network found in the Ridley 2017 research paper.

get_network_from_matrix_and_positions

Get nodes and edges from a numpy matrix and a dictionary of positions.

gnp_random_connected_graph

Create a randomly connected graph.
yawning_titan.networks.network_creator.check_if_nearby(pos, full_list, value)[source]#

Check if a randomly generated point is close to points already generated.

Parameters:

  • pos – The x,y position as a list.

  • full_list – The full list of positions.

  • value – The separation value.

Returns:

True if nearby, otherwise False.

yawning_titan.networks.network_creator.generate_node_positions(matrix)[source]#

Generate a random position for each node and saves it as a dictionary.

Parameters:

matrix – The adjacency matrix for the network.

Returns:

A dictionary of node positions.

yawning_titan.networks.network_creator.get_network_from_matrix_and_positions(matrix, positions)[source]#

Get nodes and edges from a numpy matrix and a dictionary of positions.

Parameters:

  • matrix – A 2D numpy array adjacency matrix.

  • positions – The node positions on a graph.

Returns:

An instance of Network.

yawning_titan.networks.network_creator.get_18_node_network_mesh()[source]#

The standard 18 node network found in the Ridley 2017 research paper.

Returns:

An instance of Network.

yawning_titan.networks.network_creator.dcbo_base_network()[source]#

Creates the same network used to generated DCBO data.

Returns:

An instance of Network.

New in version 1.0.1.

yawning_titan.networks.network_creator.create_mesh(size=100, connectivity=0.7)[source]#

Create a mesh node environment.

Parameters:

  • size – The number of nodes in the environment.

  • connectivity – How connected each of the nodes should be (percentage chance for any node to be connected to any other).

Returns:

An instance of Network.

yawning_titan.networks.network_creator.create_star(first_layer_size=8, group_size=5, group_connectivity=0.5)[source]#

Create a star node environment.

This is one node in the middle with groups of nodes around it. There is only one connection between a group and the center node. Groups cannot connect to each other.

Parameters:

  • first_layer_size – The number of collections of nodes in first “outer ring”.

  • group_size – How many nodes are in each collection.

  • group_connectivity – How connected the nodes in the connections are.

Returns:

An instance of Network.

yawning_titan.networks.network_creator.create_p2p(group_size=5, inter_group_connectivity=0.1, group_connectivity=1)[source]#

Create a two group network.

You can modify the connectivity between the two groups and the connectivity within the groups.

Parameters:

  • group_size – The amount of nodes in each group (before random variance).

  • inter_group_connectivity – The connectivity between the two groups.

  • group_connectivity – The connectivity within the group.

Returns:

An instance of Network.

yawning_titan.networks.network_creator.create_ring(break_probability=0.3, ring_size=60)[source]#

Create a ring network.

Parameters:

break_probability – The probability that two nodes will not be

connected.

Parameters:

ring_size – The number of nodes in the network.

Returns:

An instance of Network.

yawning_titan.networks.network_creator.custom_network()[source]#

Create custom network through user interaction.

Returns:

An instance of Network.

yawning_titan.networks.network_creator.gnp_random_connected_graph(n_nodes, probability_of_edge)[source]#

Create a randomly connected graph.

With the guarantee that each node will have at least one connection.

This is taken from the following stack overflow Q&A with a bit of a refactor for clarity.

Parameters:

  • n_nodes – the number of nodes in the graph.

  • probability_of_edge – the probability for a node to have an edge.

Returns:

An instance of Network.