# Reproducibility across machines

**URL:** <https://discourse.pymc.io/t/reproducibility-across-machines/2591>\
**Category:** Questions\
**Created:** [January 27, 2019, 4:17am UTC](https://discourse.pymc.io/t/reproducibility-across-machines/2591 "2019-01-27T04:17:37Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![djmannion](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/djmannion/32/5711_2.png) [@djmannion](https://discourse.pymc.io/u/djmannion)\
**Post date:** [January 27, 2019, 4:17am UTC](https://discourse.pymc.io/t/reproducibility-across-machines/2591/1 "2019-01-27T04:17:37Z")

</div>

Hi all,

I am having trouble getting the sampling traces to be reproducible on different machines. While running the same code multiple times on the same machine gives identical results, running the code on a different machine gives different results. They are very close, but not identical.

They have the same versions of pymc3 (3.6), theano (1.0.4), numpy (1.16.0), and scipy (1.2.0), although have different versions of Python (3.7.2 and 3.6.7). They are both on linux (current Arch, ubuntu 18.04). They do have different compilers installed, but both have empty theano configs.

For testing, I have been running `sample` as `pm.sample(1000, nuts_kwargs={"Emax":700}, cores=4, chains=4, random_seed=[1, 2, 3, 4])`.

Could I please check that I can indeed expect the traces to be identical in this scenario? I.e. is having the same model, the same `random_seed` value, and the same `pm.sample` parameters sufficient?

Thanks for any help or advice.

---

<div class="post-metadata">

**Author:** ![lucianopaz](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/lucianopaz/32/2186_2.png) [@lucianopaz](https://discourse.pymc.io/u/lucianopaz)\
**Post date:** [January 28, 2019, 9:21am UTC](https://discourse.pymc.io/t/reproducibility-across-machines/2591/2 "2019-01-28T09:21:57Z")

</div>

It is very difficult to get exactly the same traces for different machines. In the test suite, we run [a test](https://github.com/pymc-devs/pymc3/blob/master/pymc3/tests/test_step.py#L683-L735) to check if the step methods (e.g. NUTS, metropolis, SMC, etc) give a trace that “matches” the trace we expect to see. This match is not an absolute equality match (`assert all(x == y)`) but a more relaxed match (`np.allclose`) with a tolerance that depends on the floating point precision being used. With that said, across different machines there are some small differences in the traces, and as the sampling progresses, these differences may pile up and lead to very different traces at the end.

---

<div class="post-metadata">

**Author:** ![djmannion](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/djmannion/32/5711_2.png) [@djmannion](https://discourse.pymc.io/u/djmannion)\
**Post date:** [January 28, 2019, 9:50pm UTC](https://discourse.pymc.io/t/reproducibility-across-machines/2591/3 "2019-01-28T21:50:47Z")

</div>

Great, thank you for the info. For additional context, I am creating an environment on [codeocean](http://codeocean.com/) that I can refer to in a journal article to reproduce an analysis. It sounds like it isn’t as straightforward as using random seeds to get identical results on there and on my local machine.