{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,17]],"date-time":"2025-10-17T14:32:05Z","timestamp":1760711525090,"version":"3.40.3"},"reference-count":24,"publisher":"IEEE","license":[{"start":{"date-parts":[[2024,9,23]],"date-time":"2024-09-23T00:00:00Z","timestamp":1727049600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,9,23]],"date-time":"2024-09-23T00:00:00Z","timestamp":1727049600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,9,23]]},"DOI":"10.1109\/hpec62836.2024.10938437","type":"proceedings-article","created":{"date-parts":[[2025,4,3]],"date-time":"2025-04-03T19:07:19Z","timestamp":1743707239000},"page":"1-7","source":"Crossref","is-referenced-by-count":1,"title":["ModelGauge: Inference Profiling of Deep-Learning Models"],"prefix":"10.1109","author":[{"given":"Calvin B.","family":"Gealy","sequence":"first","affiliation":[{"name":"University of Pittsburgh,Electrical and Computer Engineering,Pittsburgh,United States"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"David","family":"Langerman","sequence":"additional","affiliation":[{"name":"NSF Center for Space, High-Performance, and Resilient Computing (SHREC),Pittsburgh,United States"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alan D.","family":"George","sequence":"additional","affiliation":[{"name":"University of Pittsburgh,Electrical and Computer Engineering,Pittsburgh,United States"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","first-page":"446","article-title":"MLPerf Inference Benchmark","volume-title":"2020 ACM\/ IEEE 47th Annual International Symposium on Computer Architecture (ISCA)","author":"Reddi"},{"key":"ref2","first-page":"352","article-title":"MLPerf mobile inference benchmark: An industry-standard open-source machine learning benchmark for on-device AI","volume-title":"Proceedings of Machine Learning and Systems","volume":"4","author":"Reddi","year":"2022"},{"key":"ref3","doi-asserted-by":"crossref","first-page":"542","DOI":"10.1007\/978-3-030-36592-9_44","article-title":"DLI: Deep Learning Inference Benchmark","volume-title":"Supercomputing","volume":"1129","author":"Kustikova","year":"2019"},{"issue":"3","key":"ref4","doi-asserted-by":"crossref","first-page":"8","DOI":"10.1109\/MDAT.2021.3095215","article-title":"EDLAB: A Benchmark for Edge Deep Learning Accelerators","volume":"39","author":"Kong","year":"2022","journal-title":"IEEE Design & Test"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1145\/3485447.3512148"},{"journal-title":"ONNX developers","article-title":"ONNX: Open Neural Network Exchange","year":"2019","key":"ref6"},{"journal-title":"ONNX Runtime developers","article-title":"ONNX Runtime","year":"2021","key":"ref7"},{"journal-title":"NVIDIA","article-title":"NVIDIA TensorRT","year":"2023","key":"ref8"},{"journal-title":"Valgrind Developers","article-title":"Valgrind","year":"2023","key":"ref9"},{"volume-title":"Linux perf","year":"2023","author":"Torvalds","key":"ref10"},{"journal-title":"NVIDIA Developers","key":"ref11","article-title":"TensorRT\/samples\/trtexec at main \u2022 NVIDIA\/TensorRT"},{"journal-title":"PyTorch Developers","article-title":"Models and pretrained weights - Torchvision documentation","year":"2023","key":"ref12"},{"journal-title":"Hugging Face Developers","key":"ref13","article-title":"Hugging Face Transformers"},{"key":"ref14","first-page":"10096","article-title":"EfficientNetV2: Smaller Models and Faster Training","volume-title":"Proceedings of the 38th International Conference on Machine Learning","author":"Tan","year":"2021"},{"key":"ref15","first-page":"4510","article-title":"MobileNetV2: Inverted Residuals and Linear Bottlenecks","volume-title":"2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Sandler"},{"key":"ref16","first-page":"1314","article-title":"Searching for MobileNetV3","volume-title":"2019 IEEE\/CVF International Conference on Computer Vision (ICCV)","author":"Howard"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref18","doi-asserted-by":"crossref","first-page":"87.1","DOI":"10.5244\/C.30.87","article-title":"Wide Residual Networks","volume-title":"Procedings of the British Machine Vision Conference 2016","author":"Zagoruyko","year":"2016"},{"journal-title":"An Image is Worth 16\u00d716 Words: Transformers for Image Recognition at Scale","year":"2021","author":"Dosovitskiy","key":"ref19"},{"key":"ref20","article-title":"MobileViT: Light-weight, general-purpose, and mobile-friendly vision transformer","volume-title":"International Conference on Learning Representations","author":"Mehta","year":"2022"},{"volume-title":"PyTorch-OpCounter","year":"2022","author":"Zhu","key":"ref21"},{"key":"ref22","first-page":"1","article-title":"Beyond Floating-Point Ops: CNN Performance Prediction with Critical Datapath Length","volume-title":"2020 IEEE High Performance Extreme Computing Conference (HPEC)","author":"Langerman"},{"volume-title":"AMD EPYC\u2122 9374F","key":"ref23"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3654799"}],"event":{"name":"2024 IEEE High Performance Extreme Computing Conference (HPEC)","start":{"date-parts":[[2024,9,23]]},"location":"Wakefield, MA, USA","end":{"date-parts":[[2024,9,27]]}},"container-title":["2024 IEEE High Performance Extreme Computing Conference (HPEC)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10938401\/10938415\/10938437.pdf?arnumber=10938437","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,4]],"date-time":"2025-04-04T08:09:28Z","timestamp":1743754168000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10938437\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,9,23]]},"references-count":24,"URL":"https:\/\/doi.org\/10.1109\/hpec62836.2024.10938437","relation":{},"subject":[],"published":{"date-parts":[[2024,9,23]]}}}