Loop-Tiling Based Compiling Optimization for CNN Accelerators
- 24 October 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Abstract
With the rapid development of convolutional neural networks (CNNs), CNN accelerators have been a research focus and are continuously being developed. Loop tiling is a commonly deployed optimization method to guide hardware architecture design based on hardware resource constraints. However, the potential performance gain of loop tiling at the compiler phase is rarely considered. In this paper, a loop-tiling based compiling scheme is proposed, which could assist a large range of hardware architecture to improve the versatility and performance of inference. A network reorganization scheme is introduced which is customized to the target hardware architecture and reconstructed the network model in a more hardware-friendly manner. At the same time, the weight re-ordering is performed correspondingly to the newly generated network model to guarantee correct data access. Experimental results demonstrate that the proposed method could effectively enlarge the range of supported CNN models for CNN accelerators with various hardware configurations.Keywords
This publication has 9 references indexed in Scilit:
- Flexer: Out-of-Order Scheduling for Multi-NPUs2023
- LIPFD-NPU: Low-overhead Instruction-driven Permanent Fault Detection for Neural Processing Unit2022
- Configurable CNN Accelerator in Speech Processing based on Vector Convolution2022
- RepVGG: Making VGG-style ConvNets Great Again2021
- An Efficient and Flexible Accelerator Design for Sparse Convolutional Neural NetworksIEEE Transactions on Circuits and Systems I: Regular Papers, 2021
- Stream Processing Dual-Track CGRA for Object InferenceIEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2018
- Optimizing Loop Operation and Dataflow in FPGA Acceleration of Deep Convolutional Neural Networks2017
- Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural NetworksIEEE Journal of Solid-State Circuits, 2016
- Going Deeper with Embedded FPGA Platform for Convolutional Neural Network2016