An Optimization Technique for General Neural Network Hardware Architecture

FPGA has the several advantage of highly parallel architecture, flexibility and low power utilization and it also take part in sampling of data and in industries. Artificial intelligence is one of the vital fields to achieve high efficiency for training and apply the neural network architecture but achieving high efficiency is heavily demanded. To overcome this difficulty, industries use FPGA into Artificial intelligence field. For that we tend to execute General neural network Architecture and preliminarily validate on FPGA platform by using XILINX Software which contains high processing speed and abundant hardware resources. The monolithic integration of a large number of numerous functions on a single chip usually provides low compactness, low testing necessity, high reliability, saving cost.