Application Acceleration With FPGAs
Programmable Acceleration Cards (PACs), DCP, DLA, Software Stack, and Reference Designs
Announcements
Intel Support hours are Monday-Fridays, 8am-5pm PST, except Holidays. Thanks to our community members who provide support during our down time or before we get to your questions. We appreciate you!

Need Forum Guidance? Click here
Search our FPGA Knowledge Articles here.
422 Discussions

Develop a complete RESNET33 model on Fpga using Vhdl.

RKhaz
Beginner
393 Views

Hello,

I am working on the ZYNQ Fpga, my task is to develop a complete RESNET33 model on Fpga using Vhdl. I will not be training the model but using the pre-trained weights, hence I will be doing only the inference part. 

 

So far I have developed a single layer of CNN(Zero-padding, Convolution, ReLu, Bias, MaxPool). I need some insights from you, am I heading in the right direction?

 

Is it possible to implement on Vhdl or should i chang approach to a more easy way?

 

Please note: Input Image is a grayscale bit stream.

0 Kudos
4 Replies
HRZ
Valued Contributor II
183 Views

Even though your question is not FPGA-dependant, ZYNQ FPGAs are made by Xilinx and not Intel; hence, it is probably best to ask for help on Xilinx's forums.

JEYANTHKRI_N_Intel
183 Views

Hi,

Thanks for reaching out to us.

 As this forum is focussed on Intel AI related questions and since you are working on the ZYNQ Fpga which is made by Xilinx and not Intel, we recommend you to post the question in Xilinx's forum.

 

RKhaz
Beginner
183 Views

Hello HRZ and Jeyanth,

 

Yes you are right, Thank you.

JEYANTHKRI_N_Intel
183 Views

Hi ,

Glad to hear that the solution provided helped

 

After the case closure, you will receive a survey email. We appreciate it if you can complete this survey regarding the support you received.

Reply