Creating a Dataset for High-Performance Computing Code Translation using LLMs: A Bridge Between OpenMP Fortran and C++
Authors:
- 25 September 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Abstract
In this study, we present a novel dataset for training machine learning models translating between OpenMP Fortran and C++ code. To ensure reliability and applicability, the dataset is created from a range of representative open-source OpenMP benchmarks. It is also refined using a meticulous code similarity test. The effectiveness of our dataset is assessed using both quantitative (CodeBLEU) and qualitative (human evaluation) methods. We showcase how this dataset significantly elevates the translation competencies of large language models (LLMs). Specifically, models without prior coding knowledge experienced a boost of x 5.1 in their CodeBLEU scores, while models with some coding familiarity saw an impressive x 9.9-fold increase. The best fine-tuned model using our dataset outperforms GPT-4. It is also reaching human-level accuracy. This work underscores the immense potential of our dataset in propelling advancements in the domain of code translation for high-performance computing. The dataset is accessible at https://github.com/bin123apple/Fortran-CPP-HPC-code-translation-dataset.Keywords
All Related Versions
Funding Information
- LLNL (DE-AC52-07NA27344)
This publication has 9 references indexed in Scilit:
- Review of deep learning: concepts, CNN architectures, challenges, applications, future directionsJournal of Big Data, 2021
- DeepSpeed2020
- AI for Science: Report on the Department of Energy (DOE) Town Halls on
Artificial Intelligence (AI) for Science2020
- Survey of Methodologies, Approaches, and Challenges in Parallel Programming Using High-Performance Computing SystemsScientific Programming, 2020
- An Extended Polyhedral Model for SPMD Programs and Its Use in Static Data Race Detection2017
- Automatic Fortran to C++ conversion with FABLESource Code for Biology and Medicine, 2012
- Effective Source-to-Source Outlining to Support Whole Program Empirical Optimization2010
- Case study: an incremental approach to re-engineering a legacy FORTRAN Computational Fluid Dynamics code in C++Advances in Engineering Software, 1995
- The NAS parallel benchmarks---summary and preliminary results1991