Skip to content
No description, website, or topics provided.
Jupyter Notebook Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.gitignore
experiments.py
experiments_base.py
experiments_majority.py
graph_estimators.py
graph_generators.py
mit_exp_debug.ipynb
mit_experiment.py
mit_result_debug.ipynb
plots_paper.py
readme.md

readme.md

Code for the paper "Block-Structure Based Time-Series Models For Graph Sequences" (2017)

This codebase is for algorithms proposed in the aforementioned paper. It consists of the following python3 scripts:

  • graph_generators.py : has the synthetic/real instance generators
  • graph_estimators.py : has the proposed algorithms
  • experiments.py : calls algorithms on various instances
  • plots_paper.py : generates plots given in the paper

The easiest way to get started is to look at experiments.py and go from there.

Dependencies

The prominent are: numpy, networkx, graph-tool, pulp, matplotlib.

Please make a pull request if you spot bugs or have suggestions!

You can’t perform that action at this time.