首页>
外国专利>
Histogram-based per-layer data format selection for hardware implementation of deep neutral network
Histogram-based per-layer data format selection for hardware implementation of deep neutral network
展开▼
机译:基于直方图的每层数据格式选择,用于深度中性网络的硬件实现
展开▼
页面导航
摘要
著录项
相似文献
摘要
A histogram-based method of selecting a fixed point number format for representing a set of values input to, or output from, a layer of a Deep Neural Network (DNN). The method comprises obtaining a histogram that represents an expected distribution of the set of values of the layer, each bin of the histogram is associated with a frequency value and a representative value in a floating point number format; quantising the representative values according to each of a plurality of potential fixed point number formats; estimating, for each of the plurality of potential fixed point number formats, the total quantisation error based on the frequency values of the histogram and a distance value for each bin that is based on the quantisation of the representative value for that bin; and selecting the fixed point number format associated with the smallest estimated total quantisation error as the optimum fixed point number format for representing the set of values of the layer.
展开▼