1 code implementation • 1 Feb 2024 • Wenxin Jiang, Jerin Yasmin, Jason Jones, Nicholas Synovic, Jiashen Kuo, Nathaniel Bielanski, Yuan Tian, George K. Thiruvathukal, James C. Davis
Our analysis of this dataset provides the first summary statistics for the PTM supply chain, showing the trend of PTM development and common shortcomings of PTM package documentation.
1 code implementation • 5 Oct 2023 • Wenxin Jiang, Jason Jones, Jerin Yasmin, Nicholas Synovic, Rajeev Sashti, Sophie Chen, George K. Thiruvathukal, Yuan Tian, James C. Davis
Developing and training deep learning models is expensive, so software engineers have begun to reuse pre-trained deep learning models (PTMs) and fine-tune them for downstream tasks.
no code implementations • 2 Oct 2023 • Wenxin Jiang, Chingwo Cheung, Mingyu Kim, Heesoo Kim, George K. Thiruvathukal, James C. Davis
PTM authors should choose appropriate names for their PTMs, which would facilitate model discovery and reuse.
no code implementations • 30 Mar 2023 • Purvish Jajal, Wenxin Jiang, Arav Tewari, Erik Kocinare, Joseph Woo, Anusha Sarraf, Yung-Hsiang Lu, George K. Thiruvathukal, James C. Davis
We find that the node conversion stage of a model converter accounts for ~75% of the defects and 33% of reported failure are related to semantically incorrect models.
1 code implementation • 13 Mar 2023 • Wenxin Jiang, Vishnu Banna, Naveen Vivek, Abhinav Goel, Nicholas Synovic, George K. Thiruvathukal, James C. Davis
We describe this process as deep learning model reengineering.
no code implementations • 5 Mar 2023 • Wenxin Jiang, Nicholas Synovic, Matt Hyatt, Taylor R. Schorlemmer, Rohan Sethi, Yung-Hsiang Lu, George K. Thiruvathukal, James C. Davis
We interviewed 12 practitioners from the most popular PTM ecosystem, Hugging Face, to learn the practices and challenges of PTM reuse.
no code implementations • 5 Mar 2023 • Diego Montes, Pongpatapee Peerapatanapokin, Jeff Schultz, Chengjun Gun, Wenxin Jiang, James C. Davis
Among the top 10 discrepancies, we find differences of 1. 23%-2. 62% in accuracy and 9%-131% in latency.
no code implementations • 9 Dec 2021 • Wei Ju, Wenxin Jiang
We comment on a recent TKDE paper "Linear Approximation of F-measure for the Performance Evaluation of Classification Algorithms on Imbalanced Data Sets", and make two improvements related to comparison of F-measures for two prediction rules.
1 code implementation • 2 Jul 2021 • Vishnu Banna, Akhil Chinnakotla, Zhengxin Yan, Anirudh Vegesana, Naveen Vivek, Kruthi Krishnappa, Wenxin Jiang, Yung-Hsiang Lu, George K. Thiruvathukal, James C. Davis
To promote best practices within the engineering community, academic institutions and Google have partnered to launch a Special Interest Group on Machine Learning Models (SIGMODELS) whose goal is to develop exemplary implementations of prominent machine learning models in community locations such as the TensorFlow Model Garden (TFMG).
no code implementations • 29 Dec 2020 • Wenxin Jiang
We provide analytic formulas for the standard error and confidence intervals for the F measures, based on a property of asymptotic normality in the large sample limit.
no code implementations • 19 Sep 2019 • Ruimin Zhu, Thanapon Noraset, Alisa Liu, Wenxin Jiang, Doug Downey
Word embeddings capture syntactic and semantic information about words.
no code implementations • 19 Jul 2016 • Ruimin Zhu, Wenxin Jiang
Community detection has been an active research area for decades.
no code implementations • 30 Jan 2013 • Wenxin Jiang, Martin A. Tanner
We investigate a class of hierarchical mixtures-of-experts (HME) models where exponential family regression models with generalized linear mean functions of the form psi(ga+fx^Tfgb) are mixed.