New tutorial section on Pynq

[thesps]

The hls4ml tutorial has a new section “part 7; deployment” on deploying your neural network on hardware, targeting a pynq-z2 board with the new VivadoAccelerator backend.

In the new tutorial, we show you how to take one of the neural networks trained in a previous section and create a bitfile (or "Overlay") that you can load onto a pynq-z2 device for fast inference.

The VivadoAccelerator backend was released with hls4ml v0.6.0 / coris with support for pynq-z2 and zcu102 boards. What other boards would you like to see supported? Let us know👇

[Fe-r-oz]

Kindly please have support for Arty A7 100T as well.

It's a low-cost medium resource board that can be used for educational purposes, so it's best for university students. More powerful than TUL PYNQ Z2 though!

[amahpour]

I'm working on an example with the Arty A7 but need help with the HLS generated code. Any chance you can help out?

[Fe-r-oz]

My apologies for the delayed response.

I shifted my efforts to PYNQ-Z2 instead of Arty A7.

[JochiSt]

Support for the Nexys4, which is also based on the XC7A100T and the ZyBo would be great.

[phucducnguyen]

it would be great if we have support for KV260 and KR260 as well. They are also good boards