# Declare an internal Poisson distribution inside a custom distribution failes

**URL:** <https://discourse.pymc.io/t/declare-an-internal-poisson-distribution-inside-a-custom-distribution-failes/1122>\
**Category:** Questions\
**Created:** [April 23, 2018, 5:45am UTC](https://discourse.pymc.io/t/declare-an-internal-poisson-distribution-inside-a-custom-distribution-failes/1122 "2018-04-23T05:45:36Z")\
**Posts on this page:** 2\
**Page:** 2

<div class="post-metadata">

**Author:** ![bhomass](https://avatars.discourse-cdn.com/v4/letter/b/5f8ce5/32.png) [@bhomass](https://discourse.pymc.io/u/bhomass)\
**Post date:** [April 30, 2018, 5:42am UTC](https://discourse.pymc.io/t/declare-an-internal-poisson-distribution-inside-a-custom-distribution-failes/1122/21 "2018-04-30T05:42:42Z")

</div>

The reverse ordered print out may be some kind of artifact. Only one logp is generated per round trip (one forward and one reverse iteration)

---

<div class="post-metadata">

**Author:** ![junpenglao](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/junpenglao/32/8_2.png) [@junpenglao](https://discourse.pymc.io/u/junpenglao)\
**Post date:** [April 30, 2018, 7:41am UTC](https://discourse.pymc.io/t/declare-an-internal-poisson-distribution-inside-a-custom-distribution-failes/1122/22 "2018-04-30T07:41:43Z")

</div>

The backward is probably the computation of gradient.

[Previous page](https://discourse.pymc.io/t/declare-an-internal-poisson-distribution-inside-a-custom-distribution-failes/1122.md?page=1)