# DPM model for clustering

**URL:** <https://discourse.edwardlib.org/t/dpm-model-for-clustering/97>\
**Category:** General\
**Created:** [April 22, 2017, 4:17pm UTC](https://discourse.edwardlib.org/t/dpm-model-for-clustering/97 "2017-04-22T16:17:19Z")\
**Posts on this page:** 1\
**Showing post:** 9

<div class="post-metadata">

**Author:** ![AngelBerihuete](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.edwardlib.org/angelberihuete/32/27_2.png) [@AngelBerihuete](https://discourse.edwardlib.org/u/AngelBerihuete)\
**Post date:** [August 25, 2017, 7:27am UTC](https://discourse.edwardlib.org/t/dpm-model-for-clustering/97/9 "2017-08-25T07:27:50Z")

</div>

Hi @emilemathieu !

Have you seen the post written by @dustin about your issue?

> [@Gaussian mixture with covariance matrix via klqp method](https://discourse.edwardlib.org/t/gaussian-mixture-with-covariance-matrix-via-klqp-method/277/2):
>
> ed.KLqp is known not to work very well with Gamma or Dirichlet variational approximations. For high-dimensional and large GMM problems, you should generally stick with Gibbs sampling or coordinate ascent VI. These are the gold standards. (We haven’t implemented the latter yet.)

Do you have any progress with DPM and Edward?

Cheers!

---

_[View the full topic](https://discourse.edwardlib.org/t/dpm-model-for-clustering/97)._