Yao Lai, Jinxin Liu, Zhentao Tang, Bin Wang, Jianye Hao, Ping Luo
The University of Hong Kong / Shanghai AI Lab / Huawei Noah's Lab / Zhejiang University / Tianjin University
ICML 2023
ChiPFormer casts the chip placement as an offline RL formulation and enables learning a transferable placement policy from fixed offline data, which can improve the placement efficiency and quality (up to 97% time and 65% HPWL decrease).
Place macros by ChiPFormer.
Chip: Adaptec3
Chip: Adaptec4
Fix macros, place standard cells in coarse.
Set macros and standard cells movable,
place all modules.
Chip: Adaptec3
Human
HPWL = 19.41e7Overlap = 0%ChiPFormer
HPWL = 13.97e7Overlap = 0%ChiPFormer: Adaptec4
Human
HPWL = 17.44e7Overlap = 0%DeepPR
HPWL = 23.40e7Overlap = 32.11%PRNet
HPWL = 23.64e7Overlap = 13.36%GraphPlace
HPWL = 25.58e7Overlap = 7.43%MaskPlace
HPWL = 22.97e7Overlap = 0%DREAMPlace
HPWL = 14.41e7Overlap = 0%Flora
HPWL = 14.30e7Overlap = 0%ChiPFormer
HPWL = 12.97e7Overlap = 0%We release the offline training data in Google Drive. (500 placement results for 12 chip circuits.)
Documents can be seen in Google Docs.
We have released our code on github.