构建首个跨工艺的电容提取数据集，助力机器学习模型研究。
CapBench: A Multi-PDK Dataset for Machine-Learning-Based Post-Layout Capacitance Extraction
- 基于开源芯片设计，跨3大工艺节点生成6万余个电容窗口数据。
- 采用高精度求解器生成标签，总电容误差仅0.64%，验证可靠。
- 提供密度图、图结构等多模态表示，适配多种神经网络研究。
我们提出CapBench，一个可复现的多工艺库（multi-PDK）电容提取数据集。数据源自开源设计，包括单核CPU、片上系统和媒体加速器，均通过14次独立OpenROAD流程在ASAP7、NanGate45和Sky130HD三个工艺节点完成布局布线。从这些版图中提取61,855个3D窗口，覆盖三种尺寸层级，支持迁移学习与可扩展性研究。高保真电容标签由先进随机游走求解器RWCap生成，并经工业标准Raphael验证，总电容平均绝对误差为0.64%。每个窗口预处理为密度图、图表示和点云。评估了10种机器学习架构，包括卷积神经网络（CNN）、点云变换器和图神经网络（GNN），结果显示CNN误差最低（1.75%），而GNN速度最快达41.4倍，但误差较大（10.2%），体现明显的精度-速度权衡。代码与数据集已开源：https://github.com/THU-numbda/CapBench。
原文摘要 · Abstract (English)
We present CapBench, a fully reproducible, multi-PDK dataset for capacitance extraction. The dataset is derived from open-source designs, including single-core CPUs, systems-on-chip, and media accelerators. All designs are fully placed and routed using 14 independent OpenROAD flow runs spanning three technology nodes: ASAP7, NanGate45, and Sky130HD. From these layouts, we extract 61,855 3D windows across three size tiers to enable transfer learning and scalability studies. High-fidelity capacitance labels are generated using RWCap, a state-of-the-art random-walk solver, and validated against the industry-standard Raphael, achieving a mean absolute error of 0.64% for total capacitance. Each window is pre-processed into density maps, graph representations, and point clouds. We evaluate 10 machine learning architectures that illustrate dataset usage and serve as baselines, including convolutional neural networks (CNNs), point cloud transformers, and graph neural networks (GNNs). CNNs demonstrate the lowest errors (1.75%), while GNNs are up to 41.4x faster but exhibit larger errors (10.2%), illustrating a clear accuracy-speed trade-off. Code and dataset are available at https://github.com/THU-numbda/CapBench.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。