Neural Networks on FPGA: Part 8: Generating weights, biases and test data

preview_player
Показать описание
Source code
Рекомендации по теме
Комментарии
Автор

Thank you sir, i've learnt a lot from your tutorial series! :)

reubengoh
Автор

I am lost here. Why do we change from vivado ( verilog ) design to python? Shouldn't we compare the amount of time, data sample size, amount of resources used all inside vivado simulation results? What is the purpose for using python here?
Thanks in advance.

xuezhizhang
Автор

sir can we used this model to our data.

rajveerlalawat
Автор

Hello Sir, this is undoubtedly the best tutorial I have ever seen for FPGA application of Neural Network... I need one help... I was trying to build my own neural network for an image processing application... I have my own training and testing images... but I don't know how to use those images to train the network... how can I format those images to make a dataset like MNIST? Your Python code loads the MNIST data set and use it.... your code has been written in such a way that it can handle the MNIST zip file... how can I use my images? Please answer me... it will be a great help... thanks a lot

md.saifulislam
Автор

hi, thanks for the good tutorial, I have a problem, I want to simulate this code with ModelSim from vivado, so I change target simulator to modelsim and launch all libraries needed but when I run simulation nothing happen and in the tcl console just showed these errors:
Error: Cannot find `include file "..rtlinclude.v" in directories:
# C:/modeltech64_2019.4/ovm-2.1.2/../verilog_src/ovm-2.1.2/src,
# ** Error: (vlog-2163) Macro `dataWidth is undefined.
Error: failed.

would you mind help me, I don't know why these happen?

anahitahosseinkhani