Open hildensia opened 9 years ago
So CHAMP seems to be an approximation of Fearnhead and Liu [1](whereas CHAMP uses MAP vs the full model evidence, among other things). Perhaps an implementation of [1] would be a good starting point?
[1]: P Fearnhead and Z Liu. Online inference for multiple changepoint problems. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 69(4):589-605, 2007.
Exactly. I think I'll have a look into that in the next days. I also met Scott at ICRA and had a short talk with him. Maybe it is easier to translate his C++ code to python.
Would be interesting to have champ implemented. See http://www.andrew.cmu.edu/user/sniekum/pubs/CPD15.pdf