This is an implementation of MMHC in python. All data sets and models are placed in the "Input" folder and the results are generated to the "Output" folder.
If you want to test your own data set, just put it in the "Input" folder and change the corresponding variable in "BN_structure_learning" file which is also an example file for running the code. If you want to evaluate the result, please provide the ground truth model with bnlearn format.