I am trying to implement a convolutional neural network on the terasic DE10 Nano development board using the OpenCL and Intel FPGA SDK for OpenCL. I have multiple layers in the CNN. The entire network is working fine in the emulation mode on PC. However, when I burn the code on the DE10 Nano SoC FPGA I get an error after the first few layers. The error occurs while creating a buffer and the error message is -
unable to create buffer.
Could you please help me solve this problem?
Any help in this regard is highly appreciated
Thanks and Regards