Published July 5, 2024 | Version 1.0
Conference paper Open

Interoperability in Deep Learning: A User Survey and Failure Analysis of ONNX Model Converters

  • 1. ROR icon Purdue University West Lafayette

Contributors

Contact person:

Project leader:

  • 1. ROR icon Purdue University West Lafayette

Description

This is the artifact for Interoperability in Deep Learning: A User Survey and Failure Analysis of ONNX Model Converters, that will be presented at the 2024 International Symposium on Software Testing and Analysis (ISSTA).

Instructions are in the README.

Please cite:

@inproceedings{jajal2024analysisfailuresrisksdeep,
      title={Interoperability in Deep Learning: A User Survey and Failure Analysis of {ONNX} Model Converters}, 
      author={Purvish Jajal and Wenxin Jiang and Arav Tewari and Erik Kocinare and Joseph Woo and Anusha Sarraf and Yung-      Hsiang Lu and George K. Thiruvathukal and James C. Davis},
      year={2024},
      booktitle={Proceedings of the 33nd ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA)},
      url={https://arxiv.org/abs/2303.17708},
}

Files

ISSTA_24__Analysis_of_Failures_in_Deep_Learning_Interoperability_A_Case_Study_in_the_ONNX_Ecosystem_appendix.pdf

Files (19.2 GB)

Name Size Download all
md5:4d7c644bbfaf79c3e18dd421d8372b58
3.2 MB Preview Download
md5:5e4346d479f7b9473fdde4534fca14bc
3.6 kB Preview Download
md5:7b353eb81e15a4fb036ec548c7cde84d
155 Bytes Preview Download
md5:d71530c0a76c6e5c607df7d3c920e44d
163.8 kB Download
md5:798464a787a20e50c1308f04337f5f4f
146.5 MB Download
md5:fcca3a152ee003ec509ba55a9f41473b
19.0 GB Download

Additional details

Funding

Cisco Systems (United States)
Google (United States)
U.S. National Science Foundation
Transfer Learning using Transformation among Models and Samples 1813935
U.S. National Science Foundation
Advancing Low-Power Computer Vision at the Edge 2107020
U.S. National Science Foundation
Advancing Low-Power Computer Vision at the Edge 2107230
U.S. National Science Foundation
Cyber Infrastructure to Enable Computer Vision Applications at the Edge Using Automated Contextual Analysis 2104319

Dates

Accepted
2024-07

Software