Embedded Files
Statistics of the datasets used in Baselines tables
HeCo + B2-Sampling code+data
(This code is a modification of the official HeCo code. All code and parameters are the same as from the official HeCo code, and we only add some modules for BBSampling.)
Usage:
Usage:
Example "run HeCo+B2-Sampling under dataset acm", execute:
python main.py acm --gpu=0
To run other datasets ("dblp", "aminer" or "freebase"):
Go into ./data, and use the following commend to generate the related files for shortest distance weighted sampling:
python shortestDis.py
Go into ./code, and use the following commend to run our model:
python main.py acm --gpu=0
(Here, "acm" need to be replaced by "dblp", "aminer" or "freebase")
1. Shortest path distance reflects the topological diversity of graphs.
1. Shortest path distance reflects the topological diversity of graphs.
The x-axis indicates the indices of shortest-path distance, and the y-axis indicates the ratio of positive pairs. We can see that the negative pairs lie in all shortest path distances. Following the shortest path distance, we can sample various negative pairs with different distances, capturing the topological diversity of the graph and measuring topological diversity from local to global.
2. Summary of representative GCL models within the proposed paradigm.
3. Convergence Analysis:
The loss curves of Amazon-Photo, Cora, and Amazon-Computers converged at the end.
4. Sensitivity Analysis (on homogeneous graphs): GCA + B2-Sampling on node classification and node clustering tasks.
Page updated
Google Sites
Report abuse