Has anyone trained a regression model using finn and if so how did you handle quantization at the input and output layers , i want to run the model in the pynq z2 board but am stuck at getting the output value to be int8 quantized while doing inference as mentioned in the finn tutorial notebooks for cybersec
Welcome to the PYNQ community!
You might find some more specific support for FINN on their GitHub discussions page.
1 Like