This is the implementation of our submitted paper "PL4XGL: A Programming Language Approach to Explainable Graph Learning." This implementation aims to reproduce the results of PL4XGL in Table 3, Figure 11 of our paper, and Figure 2 of our supplementary material.
- python >= 3.8.8
GraphClassification and NodeClassification folders include the implementation of PL4XGL for graph classification and node classification tasks, respectively.
if you want to reproduce the results of PL4XGL for the graph classification datasets (MUTAG, BBBP, and BACE datasets), move to GraphClassification folder and follow the instructions.
if you want to reproduce the results of PL4XGL for the node classification datasets (Texas, Cornell, Wisconsin, BA-Shapes, and Tree-Cycles datasets) move to GraphClassification folder and follow the instructions.