Code for Explanation vector?

Hi Marc, I came across your work on interpretable ML [here](http://www.nada.kth.se/~ann/exjobb/marc_beillevaire.pdf) and was wondering if you still had the code for generating explanation vectors? You had mentioned in the document that you had implemented it using the Metropolis-Hasting sampler. 

Thanks!

Sudhanshu