Ahmed Alfadhel

Hi @Ahmed Alfadhel,
Those of us at Digilent don't have any direct advice to offer regarding neural networking, but a cursory internet search yielded these two papers (link1, link2) that might be of interest to you.
Thanks,
JColvin

Hello,
I'm building this filter, generating a .COE file in Matlab, which I use in the FIR compiler IP. Here are two screenshots of the settings. Do you know if the difference between the two pictures, in terms of magnitude, are just a displaying fact or if there is a real amlpification involved by the FIR compiler ? If it's the case, do you know how to fix it to generate the same filter as I designed in Matlab, so without gain ? 
Kind regards, 
 
Yannick
 
 

@Ahmed Alfadhel,
One other thing to note is your scope was AC coupled in your pictures so you were "seeing" a bipolar waveform of -1.28v to +1.28v.
If you had DC coupled, the waveform on the scope would have shifted up and you would have observed the 0v to 2.5v the DA was outputting.
To get an actual bipolar output, you need the opamp level shifter/scaler.

well, by default your signal is between 0 V and Vref. The opamp circuit has a gain of 2 (range 0.. 2 VRef) but subtracts a constant VRef (range now -VRef..Vref).
It'll just shift the waveform on the scope, and double its AC magnitude.
 

@Ahmed Alfadhel
To understand what's going on, check out table 8 of the datasheet on page 15.  Basically, the DAC provides outputs between 0 and max, where 0 is mapped to zero and all ones is mapped to the max.  In other words, you should be plotting your data as unsigned.
To convert from your current twos complement representation to an unsigned representation where zero or idle is in the middle of the range, rather than on the far end, just toggle the MSB.
Dan