GEval: a Modular and Extensible Evaluation Framework for Graph Embedding Techniques

The ZIP file contains an evaluation framework to test graph embedding techniques upon Machine Learning - Classification, Regression, and Clustering - and semantic tasks - Entity Relatedness, Document Similarity, and Semantic Analogies. 

In the README file, it is listed how to run the framework, which parameters to set and how to customize them.

The framework supports both TXT and HDF5 vectors file. In the README file, the vector file formats are explained.

The Apache license applies to the provided source code. For the datasets, please check the licensing information. 
For example, for the licensing information of the classification and regression datasets, see https://dws.informatik.uni-mannheim.de/en/research/a-collection-of-benchmark-datasets-for-ml/; 
for Entity Relatedness task, see https://old.datahub.io/dataset/kore-50-nif-ner-corpus/resource/840dc999-8451-42d8-baaf-0647f1bc6a20.


Please, check on GitHub the last version of the framework, at Evaluation-Framework.