16-bit Half Precision Convolution Kernel Design for Convolution Neural Network

This is a project written in Verilog. This a deployment of 16 bit float-point convolution kernal on FPGA. The code and final report are attached.