This implementation supports python2.7 only. The dependencies are mentioned in requirements.txt file which can be installed using 'pip install -r requirements.txt' command. Algorithm takes adjacency matrix stored in form of csv as an input. Please set 'struc_file' parameter in config file to provide input. Encoder and decoder architecture are described in config file. 'num_clusters' parameters in config to number of communities. To run the algorithm execute 'python run.py'.