Hello everyone, I would like to know if you can implement a custom layer on the FPGA with OpenCL when you use the Inference Engine.
If this is possible, How could I do it?
In the documentation, I 've only read that you can implement a custom layer with CPU or GPU.
Dear García Martínez, Jorge,
Only certain large customers under an agreement with Intel (partners) will get support for OpenVino Custom Layers on FPGA. So yes OpenVino Custom Layers technically are supported on FPGA but not for end users.
Hope it helps,
Thank you very much, then I throw the following question:
How can I link my own kernel (for example: Implement "only a softmax" in opencl for FPGA) with the inference engine ?