# Structure learning of Bayesian networks

**URL:** <https://discourse.pymc.io/t/structure-learning-of-bayesian-networks/3845>\
**Category:** Questions\
**Created:** [September 10, 2019, 7:07pm UTC](https://discourse.pymc.io/t/structure-learning-of-bayesian-networks/3845 "2019-09-10T19:07:07Z")\
**Posts on this page:** 1\
**Showing post:** 11

<div class="post-metadata">

**Author:** ![Tanmoy.7989](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/tanmoy.7989/32/1825_2.png) [@Tanmoy.7989](https://discourse.pymc.io/u/Tanmoy.7989)\
**Post date:** [September 11, 2019, 3:21pm UTC](https://discourse.pymc.io/t/structure-learning-of-bayesian-networks/3845/11 "2019-09-11T15:21:30Z")

</div>

No. There will be hidden nodes in both DAGs. What I meant is there will be at least one observed node in the whole system. So both DAGs might have observed nodes, but a nonzero number of hidden nodes too.

---

_[View the full topic](https://discourse.pymc.io/t/structure-learning-of-bayesian-networks/3845)._