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

arXiv: https://arxiv.org/abs/2407.05466 · HTML: https://mlsanity.com/paper/2407.05466/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 96 references that are in the map.
- First posted on arXiv 7 Jul 2024.

## Builds on (most cited first, 25 of 25)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [On the Properties of Neural Machine Translation: Encoder-Decoder Approaches](https://mlsanity.com/paper/1409.1259/) (2014)
- [Linear Mode Connectivity and the Lottery Ticket Hypothesis](https://mlsanity.com/paper/1912.05671/) (2019)
- [Pre-Trained Models: Past, Present and Future](https://mlsanity.com/paper/2106.07139/) (2021)
- [Taxonomy of Real Faults in Deep Learning Systems](https://mlsanity.com/paper/1910.11015/) (2019)
- [Machine Learning in Python: Main developments and technology trends in data science, machine learning, and artificial intelligence](https://mlsanity.com/paper/2002.04803/) (2020)
- [Automatic Instrument Segmentation in Robot-Assisted Surgery Using Deep Learning](https://mlsanity.com/paper/1803.01207/) (2018)
- [SpArSe: Sparse Architecture Search for CNNs on Resource-Constrained Microcontrollers](https://mlsanity.com/paper/1905.12107/) (2019)
- [An Empirical Study towards Characterizing Deep Learning Development and Deployment across Different Frameworks and Platforms](https://mlsanity.com/paper/1909.06727/) (2019)
- [Understanding the Factors that Impact the Popularity of GitHub Repositories](https://mlsanity.com/paper/1606.04984/) (2016)
- [Toward Understanding Deep Learning Framework Bugs](https://mlsanity.com/paper/2203.04026/) (2022)
- [Deep Reinforcement Learning for Imbalanced Classification](https://mlsanity.com/paper/1901.01379/) (2019)
- [Gesture Recognition in Robotic Surgery: a Review](https://mlsanity.com/paper/2102.00027/) (2021)
- [Towards Training Reproducible Deep Learning Models](https://mlsanity.com/paper/2202.02326/) (2022)
- [A Comparative Study of Programming Languages in Rosetta Code](https://mlsanity.com/paper/1409.0252/) (2014)
- [Deterministic Implementations for Reproducibility in Deep Reinforcement Learning](https://mlsanity.com/paper/1809.05676/) (2018)
- [On Distributed Adaptive Optimization with Gradient Compression](https://mlsanity.com/paper/2205.05632/) (2022)
- [Comparison and Benchmarking of AI Models and Frameworks on Mobile Devices](https://mlsanity.com/paper/2005.05085/) (2020)
- [On the Replicability and Reproducibility of Deep Learning in Software Engineering](https://mlsanity.com/paper/2006.14244/) (2020)
- [What Do Developers Ask About ML Libraries? A Large-scale Study Using Stack Overflow](https://mlsanity.com/paper/1906.11940/) (2019)
- [Fast Changeset-based Bug Localization with BERT](https://mlsanity.com/paper/2112.14169/) (2021)
- [Deep Sequence Models for Text Classification Tasks](https://mlsanity.com/paper/2207.08880/) (2022)

## Built on by (most cited first, 1 of 1)

- [A Comparative Survey of PyTorch vs TensorFlow for Deep Learning: Usability, Performance, and Deployment Trade-offs](https://mlsanity.com/paper/2508.04035/) (2025)

## Abstract (arXiv)

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.