HLSFactory: A Framework Empowering High-Level Synthesis Datasets for Machine Learning and Beyond
Machine learning (ML) techniques have been applied to high-level synthesis (HLS) flows for quality-of-result (QoR) prediction and design space exploration (DSE). Nevertheless, the scarcity of accessible high-quality HLS datasets and the complexity of building such datasets present great challenges to FPGA and ML researchers. Existing datasets either cover only a subset of previously published benchmarks, provide no way to enumerate optimization design spaces, are limited to a specific vendor, or have no reproducible and extensible software for dataset construction. Many works also lack user-friendly ways to add more designs to existing datasets, limiting wider adoption and sustainability of such datasets.
Authors
- Hanqiu Chen (ORCID: https://orcid.org/0009-0007-1834-5612)
- Sean Lowe
- Rishov Sarkar (ORCID: https://orcid.org/0000-0002-9168-0392)
- Aman Arora (ORCID: https://orcid.org/0000-0003-2547-4424)
- Zhigang Wei (ORCID: https://orcid.org/0000-0002-2984-2098)
- Cong Hao (ORCID: https://orcid.org/0000-0002-2541-8767)
- Lizy K. John (ORCID: https://orcid.org/0000-0002-8747-5214)
- Stefan Abi-Karam (ORCID: https://orcid.org/0000-0002-6697-8517)
- Allison Seigler
- Nanditha Rao (ORCID: https://orcid.org/0000-0003-2369-0836)
Institutions
- Georgia Institute of Technology (US)
- International Institute of Information Technology Bangalore (IN)
- Georgia Tech Research Institute
- Arizona State University (US)
- The University of Texas at Austin (US)
Publication Details
- Published
- 2024-09-03
- DOI
- https://doi.org/10.1145/3670474.3685961
- Citations
- 1
- Primary Topic
- Parallel Computing and Optimization Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.49