Precise congestion prediction from a placement solution plays a crucial role in circuit placement. This work proposes the lattice hypergraph (LH-graph), a novel graph formulation for circuits, which preserves netlist data during the whole learning process, and enables the congestion information propagated geometrically and topologically. Based on the formulation, we further developed a heterogeneous graph neural network architecture LHNN, jointing the routing demand regression to support the congestion spot classification. LHNN constantly achieves more than 35% improvements compared with U-nets and Pix2Pix on the F1 score. We expect our work shall highlight essential procedures using machine learning for congestion prediction.
@article{arxiv.2203.12831,
title = {LHNN: Lattice Hypergraph Neural Network for VLSI Congestion Prediction},
author = {Bowen Wang and Guibao Shen and Dong Li and Jianye Hao and Wulong Liu and Yu Huang and Hongzhong Wu and Yibo Lin and Guangyong Chen and Pheng Ann Heng},
journal= {arXiv preprint arXiv:2203.12831},
year = {2022}
}
Comments
Accepted as a conference paper in DAC 2022; 6 pages, 4 figures