[mlpack-git] [mlpack/mlpack] Adds a GammaDistribution object for estimating gamma parameters (#729)

[Ryan Curtin]

Thanks for taking the time to figure out the Boost backporting situation.  The tests look good and the implementation seems good too (I think maybe there are a few small speedups we can do).

Do you have time to implement the other functions used by other distributions?  Specifically it would be `Probability(arma::vec&)`, `Random()`, `Train(observations, probabilities)` (that's the hard one, since we have to factor in different weights for each point; at that point we'd be deriving an original algorithm for estimation, but I think it might be a fairly straightforward modification), `Serialize()`, and `LogProbability()` (plus tests).  If you don't, that's fine, but then I would ask if you can instead just open an issue mentioning that each of these methods needs to be implemented for `GammaDistribution` to be a true distribution that can be used in the rest of mlpack. :)

---
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/mlpack/mlpack/pull/729#issuecomment-234445008
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack-git/attachments/20160721/3f6f58dc/attachment.html>