Analysis of Failures and Risks in Deep Learning Model Converters: A Case Study in the ONNX Ecosystem

arXiv:2303.17708v3 Announce Type: replace-cross
Abstract: Software engineers develop, fine-tune, and deploy deep learning (DL) models using a variety of development frameworks and runtime environments. DL model converters move models between frameworks and to runtime environments. Conversion errors compromise model quality and disrupt deployment. However, the failure characteristics of DL model converters are unknown, adding risk when using DL interoperability technologies.
This paper analyzes failures in DL model converters. We survey software engineers about DL interoperability tools, use cases, and pain …

abstract analysis arxiv case case study conversion cs.lg cs.se deep learning deploy deployment development disrupt ecosystem engineers environments errors frameworks onnx quality risks software software engineers study type