# Accelarating Convolution Neural Networks on PYNQ-Z1

**URL:** <https://discuss.pynq.io/t/accelarating-convolution-neural-networks-on-pynq-z1/1096>\
**Category:** Support\
**Created:** [April 3, 2020, 11:40am UTC](https://discuss.pynq.io/t/accelarating-convolution-neural-networks-on-pynq-z1/1096 "2020-04-03T11:40:46Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![hobbyist](https://avatars.discourse-cdn.com/v4/letter/h/5f8ce5/32.png) [@hobbyist](https://discuss.pynq.io/u/hobbyist)\
**Post date:** [April 3, 2020, 11:40am UTC](https://discuss.pynq.io/t/accelarating-convolution-neural-networks-on-pynq-z1/1096/1 "2020-04-03T11:40:46Z")

</div>

Hello, in order to accelarate CNN, should I just run my python code to the terminal of PYNQ-Z1 or should I transform it to HDL with Vivado and HDL (VHDL, Verilog)??

Thanks…

---

<div class="post-metadata">

**Author:** ![cathalmccabe](https://sea1.discourse-cdn.com/flex019/user_avatar/discuss.pynq.io/cathalmccabe/32/3348_2.png) [@cathalmccabe](https://discuss.pynq.io/u/cathalmccabe)\
**Post date:** [April 3, 2020, 12:03pm UTC](https://discuss.pynq.io/t/accelarating-convolution-neural-networks-on-pynq-z1/1096/2 "2020-04-03T12:03:28Z")

</div>

Yes, you need to create the hardware design. Running (Python/C) software in the terminal will only execute that software on the ARM processor.  
You could create the design using VHDL/Verilog, or HLS (C based), or other design methods.

Cathal

---

<div class="post-metadata">

**Author:** ![hobbyist](https://avatars.discourse-cdn.com/v4/letter/h/5f8ce5/32.png) [@hobbyist](https://discuss.pynq.io/u/hobbyist)\
**Post date:** [July 10, 2020, 8:32am UTC](https://discuss.pynq.io/t/accelarating-convolution-neural-networks-on-pynq-z1/1096/3 "2020-07-10T08:32:43Z")

</div>

Do you follow any tutorial for this design? I mean how do I create a hardware design by using C in order to running it to the FPGA (not to the CPU)?