Connectivity-Determination-Algorithm-for-Complex-Directed-Networks

Results Reproduction Prerequisite

Code Download

git clone https://github.com/zzy1130/Connectivity-Determination-Algorithm-for-Complex-Directed-Networks.git

Environment Suggestions

python==3.8.20
numpy==1.24.3
networkx==2.8.4
matplotlib==3.7.1
pandas==1.5.3
tqdm==4.65.0

Use Cases

A. Adjacency matrix input

In main.py, replace the content of input_string variable at line 65 with your matrix and run (the sample matrix has been given):

python main.py

After running, the connectivity result will be given.

B. A single random network

Example usage:

python main.py --random_net True --num_nodes 2000 --num_edges 4000

num_nodes specifies the number of nodes of the network and num_edges specifies that of edges. After running, directed_network.jpg will be outputted to the main directory. nodes.csv and edges.csv will be generated, indicating the node set and edge set of the network. reduced_nodes.csv and reduced_edges.csv will be generated, indicating the node set and edge set of the reduced network after running tarjan algorithm. In addition, the connectivity type will be printed in the output.

C. Enumerate all networks with node number and edge number being specified

Example usage:

python random_network.py --constrained True --num_nodes 5 --num_edges 5

for injective graphs

python random_network.py --num_nodes 5 --num_edges 5

for graphs with no constraints

D. The Erd\H{o}s--R'{e}nyi graphs experiment

Example usage:

python erdos_renyi.py --num_nodes 1000 --start_p 0.01 --final_p 0.5 --sample_num 1000 --p_num 51

num_nodes specifies the number of nodes inside the network, start_p is the initial value of p, final_p is the final value of p, sample_num specifies the number of random samples generated for each p value, and p_num is the number of p value in the experiment.