Automated Detection of Inter-Language Design Smells in Multi-Language Deep Learning Frameworks

TL;DR

The paper tackles the maintainability challenges arising from inter-language design smells (ILDS) in multi-language deep learning frameworks that use Python and C/C++. It defines seven ILDS, develops detection rules based on three inter-language communication mechanisms, and implements the CPsmell tool to automatically identify ILDS across five major DLFs with an overall validation accuracy of 98.17%. An empirical study across 50 versions reveals that ILDS are prevalent and tend to rise over time, with LLF and UNE being particularly common and certain smells like EILC and LILBC showing notable fixability in some frameworks. The work provides actionable guidance for practitioners and a roadmap for extending ILDS analysis to additional language pairs and frameworks, ultimately contributing to improved maintainability of multi-language DL systems.

Abstract

Nowadays, most DL frameworks (DLFs) use multilingual programming of Python and C/C++, facilitating the flexibility and performance of the DLF. However, inappropriate interlanguage interaction may introduce design smells involving multiple programming languages (PLs), i.e., Inter-Language Design Smells (ILDS). Despite the negative impact of ILDS on multi-language DLFs, there is a lack of an automated approach for detecting ILDS in multi-language DLFs and a comprehensive understanding on ILDS in such DLFs. This work automatically detects ILDS in multi-language DLFs written in the combination of Python and C/C++, and to obtain a understanding on such ILDS in DLFs. We first developed an approach to automatically detecting ILDS in the multi-language DLFs written in the combination of Python and C/C++, including a number of ILDS and their detection rules defined based on inter-language communication mechanisms and code analysis. We then developed the CPSMELL tool that implements detection rules for automatically detecting such ILDS, and manually validated the accuracy of the tool. Finally, we performed a study to evaluate the ILDS in multi-language DLFs. We proposed seven ILDS and achieved an accuracy of 98.17% in the manual validation of CPSMELL in 5 popular multi-language DLFs. The study results revealed that among the 5 DLFs, TensorFlow, PyTorch, and PaddlePaddle exhibit relatively high prevalence of ILDS; each smelly file contains around 5 ILDS instances on average, with ILDS Long Lambda Function For Inter-language Binding and Unused Native Entity being relatively prominent; throughout the evolution process of the 5 DLFs, some ILDS were resolved to a certain extent, but the overall count of ILDS instances shows an upward trend. The automated detection of the proposed ILDS achieved a high accuracy, and the study provides a comprehensive understanding on ILDS in the multi-language DLFs.

Figures (5)

• Figure 1: Inter-language design smell extraction process
• Figure 2: Process of creating an extension module using the Python/C API
• Figure 3: An example of design smell Excessive Inter-Language Communication
• Figure 4: Execution process of the CPsmell tool
• Figure 5: The evolution of ILDS in the 5 DLFs (RQ3)