# Tensorflow and Keras on Pynq Z1

**URL:** <https://discuss.pynq.io/t/tensorflow-and-keras-on-pynq-z1/305>\
**Category:** Support\
**Created:** [July 3, 2019, 3:42pm UTC](https://discuss.pynq.io/t/tensorflow-and-keras-on-pynq-z1/305 "2019-07-03T15:42:50Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![alkalir](https://sea1.discourse-cdn.com/flex019/user_avatar/discuss.pynq.io/alkalir/32/201_2.png) [@alkalir](https://discuss.pynq.io/u/alkalir)\
**Post date:** [July 3, 2019, 3:42pm UTC](https://discuss.pynq.io/t/tensorflow-and-keras-on-pynq-z1/305/1 "2019-07-03T15:42:50Z")

</div>

Hi,  
is there anybody that successfully built Tensorflow and Keras libraries for Pynq Z1?

Thank you.

---

<div class="post-metadata">

**Author:** ![rock](https://sea1.discourse-cdn.com/flex019/user_avatar/discuss.pynq.io/rock/32/32_2.png) [@rock](https://discuss.pynq.io/u/rock)\
**Post date:** [July 3, 2019, 4:11pm UTC](https://discuss.pynq.io/t/tensorflow-and-keras-on-pynq-z1/305/2 "2019-07-03T16:11:13Z")

</div>

I tried it before. You need to use QEMU to cross compile but there seems to be a problem with the bootstrapping of bazel. I have heard someone did it before with older versions of pynq and tensorflow, by compile directly on board; but I had no success with bazel later than 0.5.3.

I have found something I did here: [Build Bazel on QEMU for Pynq-Z1/Pynq-Z2 · GitHub](https://gist.github.com/yunqu/87568921ba2758b87dd0ec57f301efaf)