用于可重构数据流硬件布局的学习型成本模型
Learned Cost Model for Placement on Reconfigurable Dataflow Hardware
AI总结:
针对ML模型数据流图映射到可重构系统的吞吐量评估难题,提出一种学习型成本模型,相比手工分析模型准确度提升31%-52%,且编译图速度提升5.6%。
摘要:
将ML模型的数据流图映射到可重构系统是困难的,因为不同映射具有不同吞吐量并以不同方式消耗资源约束。为解决此问题,由于完全测量吞吐量成本高昂,需要一个评估映射吞吐量的模型。许多人使用手工设计的分析模型,依赖代理特征或直觉,引入了误差。我们提供了一种学习方法,在多种图上预测吞吐量的准确度提高了31%-52%。此外,该方法在移除性能注释后未出现准确度下降。使用该方法编译的图速度提升了5.6%。
英文摘要:
Mapping a dataflow-graph of an ML model onto a reconfigurable system is difficult, as different mappings have different throughputs and consume resource constraints differently. To solve this, a model to evaluate the throughput of mappings is necessary as measuring throughput completely is expensive. Many use a hand-designed analytical model, relying on proxy features or intuition, introducing error. We provide a Learned Approach that predicts throughput 31%-52% more accurately over a variety of graphs. In addition, our approach shows no accuracy degradation after removing performance annotations. We show that using this approach results in 5.6% faster compiled graphs.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。