Dataflow/mapping decides the compute and energy efficiency of DNN accelerators. Many mappers have been proposed to tackle the intra-layer map-space. However, mappers for inter-layer map-space (aka layer-fusion map-space), have been rarely discussed. In this work, we propose a mapper, DNNFuser, specifically focusing on this layer-fusion map-space. While existing SOTA DNN mapping explorations rely on search-based mappers, this is the first work, to the best of our knowledge, to propose a one-shot inference-based mapper. We leverage Transformer as our DNN architecture to learn layer-fusion optimization as a sequence modeling problem. Further, the trained DNNFuser can generalize its knowledge and infer new solutions for unseen conditions. Within one inference pass, DNNFuser can infer solutions with compatible performance to the ones found by a highly optimized search-based mapper while being 66x-127x faster.
翻译:数据流/映射决定 DNN 加速器的计算和能效。 许多映射器已被提议要处理内部映射空间。 然而, 平层间映射空间( 层融合映射空间) 的映射器很少讨论 。 在这项工作中, 我们提议一个映射器 DNNFuser, 具体侧重于该层融合映射空间 。 虽然现有的 SOTA DNN 映射勘探依靠搜索定位器, 但根据我们所知, 这是首次提出一个以一枪为根据的推断映射器。 我们利用变换器作为我们的 DNNN 结构来学习以层融合优化为序列模型的问题 。 此外, 受过训练的 DNNNF 用户可以将其知识概括化, 并推导出新的未知条件解决方案 。 在一条推论中, DNNNF 用户可以推导出与高度优化的搜索映射器所找到的兼容性能的解决方案, 同时速度为66x-127x 。