The file provides the implmentations of the algorithms(learning-based and the MVV) for the adjacency list model in the experiments in the paper "Triangle and Four Cycle Counting with Predictions in Graph Streams", which are the simulative and deterministic space version that simulate the process of sampling the edges, for the purpose of faster running time and easy to compare.

To run the code:

1. run the "compute_edge.py", which will compute R_xy and the predictions for each edges(In the Oregon graph, the predictor is the N_xy in graph#1, and then we only save 10% of the edges for predictions), based on a fixed random vertex arrival order.

2 run the "sim_MVV.py(we set the eps = 0.1)", "sim_ourAlg.py" or "sim_multi_layer.py".

The sample dataset is the Oregon dataset("graph1.txt" and "graph2.txt").

For more information of the other datasets, we can see the file "README.md" in the file for the arbitrary order experiments.