EvalNE - A Framework for Evaluating Network Embeddings

EvalNE is an open source Python library designed for assessing and comparing the performance of Network Embedding (NE) methods on Link Prediction (LP), Sign Prediction (SP), Network Reconstruction (NR), Node Classification (NC) and vizualization downstream tasks. The library intends to simplify these complex and time consuming evaluation processes by providing automation and abstraction of tasks such as model hyper-parameter tuning and model validation, node and edge sampling, node-pair embedding computation, results reporting and data visualization, among many others. EvalNE can be used both as a command line tool and as an API and is compatible with Python 3. In its current version, EvalNE can evaluate weighted directed and undirected simple networks.

EvalNE is provided under the MIT free software licence and is maintained by Alexandru Mara (alexandru(dot)mara(at)ugent(dot)be). The source code can be found on GitHub.

See the quickstart to get started.

Getting started

• Features
  • For Methodologists
  • For Practitioners
• Installation
  • Linux/MacOS
• Quickstart
  • As a command line tool
  • As an API
  • Output
  • Parallelization

Developer

• API
  • evalne package
• Release Log
  • EvalNE v0.4.0
  • EvalNE v0.3.4
  • EvalNE v0.3.3
  • EvalNE v0.3.2
  • EvalNE v0.3.1
  • EvalNE v0.3.0
  • EvalNE v0.2.3
  • EvalNE v0.2.2
  • EvalNE v0.2.1
  • EvalNE v0.2.0
• Contributing

More

• License
• Acknowledgements
• Help

Citation

If you have found EvaNE useful in your research, please cite our arXiv paper :

@article{MARA2022evalne,
  title = {EvalNE: A Framework for Network Embedding Evaluation},
  author = {Alexandru Mara and Jefrey Lijffijt and Tijl {De Bie}},
  journal = {SoftwareX},
  volume = {17},
  pages = {},
  year = {2022},
  issn = {100997},
  doi = {10.1016/j.softx.2022.100997},
  url = {https://www.sciencedirect.com/science/article/pii/S2352711022000139}
}