arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.05466cs.SEcs.AI
收藏

研究TensorFlow与PyTorch绑定对机器学习软件质量的影响

Studying the Impact of TensorFlow and PyTorch Bindings on Machine Learning Software Quality

University of Alberta(阿尔伯塔大学);Centre for Software Excellence, Huawei Canada(华为加拿大软件卓越中心)

Hao Li, Gopi Krishnan Rajbahadur, Cor-Paul Bezemer

AI总结:

该研究针对TensorFlow与PyTorch的多语言绑定,在五种深度学习模型上验证了跨绑定训练推理无准确率损失,且非默认绑定可在保精度的同时降低时间成本、提升软件质量。

摘要:

机器学习框架(如TensorFlow和PyTorch)的绑定机制允许开发者使用框架默认语言(通常是Python)之外的编程语言来集成框架功能。本文研究了在C#、Rust、Python和JavaScript中使用TensorFlow与PyTorch绑定时,在五种广泛使用的深度学习模型上进行训练和推理的软件质量影响,质量指标涵盖正确性(训练与测试准确率)和时间成本(训练与推理时间)。实验结果表明,同一框架下,模型可以在一种绑定环境中训练,再在另一种绑定环境中进行推理,且不会损失准确率。本研究首次证实,与默认的Python绑定相比,使用非默认绑定可在保持同等正确性水平的同时,从时间成本角度提升机器学习软件的质量。

英文摘要:

Bindings for machine learning frameworks (such as TensorFlow and PyTorch) allow developers to integrate a framework's functionality using a programming language different from the framework's default language (usually Python). In this paper, we study the impact of using TensorFlow and PyTorch bindings in C#, Rust, Python and JavaScript on the software quality in terms of correctness (training and test accuracy) and time cost (training and inference time) when training and performing inference on five widely used deep learning models. Our experiments show that a model can be trained in one binding and used for inference in another binding for the same framework without losing accuracy. Our study is the first to show that using a non-default binding can help improve machine learning software quality from the time cost perspective compared to the default Python binding while still achieving the same level of correctness.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑