# Testing Tensorflow 2.5 in Zybo-Z7 running Pynq 2.7

**URL:** <https://discuss.pynq.io/t/testing-tensorflow-2-5-in-zybo-z7-running-pynq-2-7/4157>\
**Category:** Learn\
**Created:** [May 6, 2022, 9:37pm UTC](https://discuss.pynq.io/t/testing-tensorflow-2-5-in-zybo-z7-running-pynq-2-7/4157 "2022-05-06T21:37:40Z")\
**Posts on this page:** 1\
**Showing post:** 10

<div class="post-metadata">

**Author:** ![dorfell](https://avatars.discourse-cdn.com/v4/letter/d/ba8739/32.png) [@dorfell](https://discuss.pynq.io/u/dorfell)\
**Post date:** [September 30, 2023, 3:10am UTC](https://discuss.pynq.io/t/testing-tensorflow-2-5-in-zybo-z7-running-pynq-2-7/4157/10 "2023-09-30T03:10:59Z")

</div>

Hello, Nihat. I hope you are doing well. When I installed TensorFlow in the Zybo-Z7 I used the last compiled binary available for ARM 32-bits.  
However, this binary creates an error when using the _model.fit_ method that does not allow training the model.  
If you want to use a machine learning framework in a Pynq-supported board with a 32-bit ARM processor, I would suggest using PyTorch (see  
[https://discuss.pynq.io/t/testing-pytorch-1-8-in-zybo-z7-running-pynq-2-7/4181](https://discuss.pynq.io/t/testing-pytorch-1-8-in-zybo-z7-running-pynq-2-7/4181)).

Please let me know if this is useful.  
Regards,  
Dorfell 🙂

---

_[View the full topic](https://discuss.pynq.io/t/testing-tensorflow-2-5-in-zybo-z7-running-pynq-2-7/4157)._