M2 project for GNN

Belief Propagation and its variants (loopy BP, uniformly reweighted BP (URW))

• The algorithm (and some useful function) are implemented in bp.py

• We test the methods (BP and loopy BP) with the exemple from here

• We control that there working well in the test_image.ipynb (using a grey scale test image)

• We measure the performance of our algorithm on (some) cow images of the TU Darmstad dataset in test_dataset.ipynb (Download/extract the dataset and place the TUDarmstadt folder at the root of the project to run this file)