Embarking on the challenging expedition from For-tran's venerable era to the dynamic landscape of Python presents a formidable task in the world of legacy code. This case study intricately navigates the transformative journey of translating a 1970s software, marked by scant documentation and challenging maintainability, into Python, employing the prowess of Large Language Model (LLM). Through a detailed exploration of the encountered nuances, pitfalls, and triumphs, this article vividly illustrates a real-world application of state-of-the-art language models in the realm of software modernization. Our experi-mentation with cutting-edge LLMs on authentic code reveals that while challenges persist, these models serve as invaluable tools for expediting the modernization process. This endeavor not only promises to breathe new life into aged software but also underscores its tangible societal and practical impact across top-tier industries.
Bridging Eras: Transforming Fortran Legacies into Python with the Power of Large Language Models / Pietrini, R.; Paolanti, M.; Frontoni, E.. - 31:(2024), pp. 1-5. (Intervento presentato al convegno 3rd IEEE International Conference on Computing and Machine Intelligence, ICMI 2024 tenutosi a usa nel 2024) [10.1109/ICMI60790.2024.10586058].
Bridging Eras: Transforming Fortran Legacies into Python with the Power of Large Language Models
Pietrini R.
Primo
;Paolanti M.Secondo
;Frontoni E.Ultimo
2024-01-01
Abstract
Embarking on the challenging expedition from For-tran's venerable era to the dynamic landscape of Python presents a formidable task in the world of legacy code. This case study intricately navigates the transformative journey of translating a 1970s software, marked by scant documentation and challenging maintainability, into Python, employing the prowess of Large Language Model (LLM). Through a detailed exploration of the encountered nuances, pitfalls, and triumphs, this article vividly illustrates a real-world application of state-of-the-art language models in the realm of software modernization. Our experi-mentation with cutting-edge LLMs on authentic code reveals that while challenges persist, these models serve as invaluable tools for expediting the modernization process. This endeavor not only promises to breathe new life into aged software but also underscores its tangible societal and practical impact across top-tier industries.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
