This repository provides a reference implementation of node2vec as described in the paper:
node2vec: Scalable Feature Learning for Networks.
Aditya Grover and Jure Leskovec.
Knowledge Discovery and Data Mining, 2016.
The node2vec algorithm learns continuous representations for nodes in any (un)directed, (un)weighted graph. Please check the project page for more details.
To run node2vec on Zachary's karate club network, execute the following command from the project home directory:
python src/main.py --input graph/karate.edgelist --output emb/karate.emd
You can check out the other options available to use with node2vec using:
python src/main.py --help
The supported input format is an edgelist:
node1_id_int node2_id_int <weight_float, optional>
The graph is assumed to be undirected and unweighted by default. These options can be changed by setting the appropriate flags.
The output file has n+1 lines for a graph with n vertices. The first line has the following format:
num_of_nodes dim_of_representation
The next n lines are as follows:
node_id dim1 dim2 ... dimd
where dim1, ... , dimd is the d-dimensional representation learned by node2vec.
If you find node2vec useful for your research, please consider citing the following paper:
@inproceedings{node2vec-kdd2016,
author = {Grover, Aditya and Leskovec, Jure},
title = {node2vec: Scalable Feature Learning for Networks},
booktitle = {Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
year = {2016}
}
Please send any questions you might have about the code and/or the algorithm to adityag@cs.stanford.edu.
Note: This is only a reference implementation of the node2vec algorithm and could benefit from several performance enhancement schemes, some of which are discussed in the paper.