{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,18]],"date-time":"2026-06-18T04:11:14Z","timestamp":1781755874938,"version":"3.54.5"},"reference-count":73,"publisher":"Springer Science and Business Media LLC","issue":"4","license":[{"start":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T00:00:00Z","timestamp":1743465600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,4,2]],"date-time":"2025-04-02T00:00:00Z","timestamp":1743552000000},"content-version":"vor","delay-in-days":1,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100014374","name":"Universitat Polit\u00e8cnica de Catalunya","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100014374","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Computing"],"published-print":{"date-parts":[[2025,4]]},"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>Machine Learning (ML)-based solutions have currently surpassed human performance in tasks like image classification, visual reasoning, and English understanding. However, this advancement comes at the cost of increasing energy consumption. Traditionally, ML projects have prioritized accuracy over energy, creating a gap in energy consumption during model inference. This study aims to (i) understand image classification datasets and pre-trained models, which is essential for the subsequent analyses, (ii) improve inference efficiency by comparing optimized and non-optimized models, and (iii) assess the economic impact of the optimizations. We conduct a controlled experiment to evaluate the impact of various PyTorch optimization techniques (dynamic quantization, , local pruning, and global pruning) on 42 Hugging Face models for image classification. The metrics examined include GPU utilization, power and energy consumption, accuracy, time, computational complexity, and economic costs. The models are repeatedly evaluated to quantify the effects of these optimization techniques. Dynamic quantization demonstrates inference time and energy consumption reductions, making it suitable for large-scale systems.  balances accuracy and energy. Local pruning shows no positive impact on accuracy, and global pruning\u2019s longer optimization times impact economic costs. This study highlights the role of software engineering tactics in achieving greener ML models, offering guidelines for practitioners to make informed decisions on optimization methods that align with sustainability goals.<\/jats:p>","DOI":"10.1007\/s00607-025-01437-8","type":"journal-article","created":{"date-parts":[[2025,4,4]],"date-time":"2025-04-04T11:26:33Z","timestamp":1743765993000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["Impact of ML optimization tactics on greener pre-trained ML models"],"prefix":"10.1007","volume":"107","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-7634-0343","authenticated-orcid":false,"given":"Alexandra","family":"Gonz\u00e1lez","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-6385-9255","authenticated-orcid":false,"given":"Joel","family":"Casta\u00f1o","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9733-8830","authenticated-orcid":false,"given":"Xavier","family":"Franch","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9928-133X","authenticated-orcid":false,"given":"Silverio","family":"Mart\u00ednez-Fern\u00e1ndez","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,4,2]]},"reference":[{"key":"1437_CR1","unstructured":"(2016) NVIDIA - developer.nvidia.com. https:\/\/developer.download.nvidia.com\/compute\/DCGM\/docs\/nvidia-smi-367.38.pdf. Accessed 30 April 2024"},{"key":"1437_CR2","unstructured":"(2021) netio-products.com. https:\/\/www.netio-products.com\/files\/download\/sw\/version\/PowerBOX-xxx-MANUAL-en_3-2-6.pdf. Accessed 30 April 2024"},{"key":"1437_CR3","unstructured":"(2024) Artificial Intelligence Index Report 2024. https:\/\/aiindex.stanford.edu\/report\/. Accessed 04 July 2024"},{"key":"1437_CR4","unstructured":"(2024) AWS Product and Service Pricing | Amazon Web Services. https:\/\/calculator.aws\/#\/createCalculator\/ec2-enhancement?nc2=pr. Accessed 14 July 2024"},{"key":"1437_CR5","unstructured":"(2024a) Introduction to model optimization for deployment - Community Computer Vision Course. https:\/\/huggingface.co\/learn\/computer-vision-course\/unit9\/intro_to_model_optimization. Accessed 14 May 2024"},{"key":"1437_CR6","unstructured":"(2024b) Optimize inference using torch.compile(). https:\/\/huggingface.co\/docs\/transformers\/perf_torch_compile. Accessed 14 May 2024"},{"key":"1437_CR7","unstructured":"(2024) Software Carbon Intensity (SCI). https:\/\/sci.greensoftware.foundation,. Accessed 13 June 2024"},{"key":"1437_CR8","unstructured":"(n.d.a) HfApi Client. https:\/\/huggingface.co\/docs\/huggingface_hub\/package_reference\/hf_api#huggingface_hub.HfApi. Accessed 13 July 2024"},{"key":"1437_CR9","unstructured":"(n.d.) PyTorch Quantization \u2013 pytorch.org. https:\/\/pytorch.org\/docs\/stable\/quantization.html. Accessed 30 April 2024"},{"key":"1437_CR10","unstructured":"(n.d.b) Tasks - Hugging Face. https:\/\/huggingface.co\/tasks. Accessed 20 July 2024"},{"key":"1437_CR11","unstructured":"(n.d.) TensorFlow Model Optimization. https:\/\/www.tensorflow.org\/model_optimization. Accessed 15 July 2024"},{"key":"1437_CR12","doi-asserted-by":"publisher","unstructured":"Adnan M, Ba Q, Shaikh N, et\u00a0al (2024) Structured model pruning for efficient inference in computational pathology. CoRR arXiv:2404.08831. https:\/\/doi.org\/10.48550\/ARXIV.2404.08831","DOI":"10.48550\/ARXIV.2404.08831"},{"key":"1437_CR13","doi-asserted-by":"crossref","unstructured":"Ajel S, Ribeiro F, Ejbali R, et\u00a0al (2022) Energy efficiency of python machine learning frameworks. ISDA. Springer, pp 586\u2013595","DOI":"10.1007\/978-3-031-35507-3_57"},{"key":"1437_CR14","doi-asserted-by":"crossref","unstructured":"Bossard L, Guillaumin M, Van\u00a0Gool L (2014) Food-101\u2013Mining Discriminative Components with Random Forests. In: ECCV 2014. Springer, pp 446\u2013461","DOI":"10.1007\/978-3-319-10599-4_29"},{"key":"1437_CR15","unstructured":"Caldiera VRBG, Rombach HD (1994) The Goal Question Metric approach. Encyclopedia of software engineering, pp 528\u2013532"},{"key":"1437_CR16","doi-asserted-by":"publisher","unstructured":"Casta\u00f1o J, Mart\u00ednez-Fern\u00e1ndez S, Franch X (2024) Lessons learned from mining the hugging face repository. In: IEEE\/ACM WSESE, WSESE\u201924, p 1\u20136. https:\/\/doi.org\/10.1145\/3643664.3648204","DOI":"10.1145\/3643664.3648204"},{"key":"1437_CR17","unstructured":"Chariton A (2024) Knowledge distillation tutorial. https:\/\/pytorch.org\/tutorials\/beginner\/knowledge_distillation_tutorial.html. Accessed 09 May 2024"},{"key":"1437_CR18","doi-asserted-by":"publisher","unstructured":"Cheng H, Zhang M, Shi JQ (2024) A survey on deep neural network pruning: taxonomy, comparison, analysis, and recommendations. IEEE TPAMI, pp 1\u20132. https:\/\/doi.org\/10.1109\/TPAMI.2024.3447085","DOI":"10.1109\/TPAMI.2024.3447085"},{"key":"1437_CR19","doi-asserted-by":"publisher","DOI":"10.4324\/9780203771587","volume-title":"Statistical power analysis for the behavioral sciences","author":"J Cohen","year":"2013","unstructured":"Cohen J (2013) Statistical power analysis for the behavioral sciences. Routledge, London"},{"key":"1437_CR20","doi-asserted-by":"publisher","unstructured":"Courty B, Schmidt V, Luccioni S, et\u00a0al (2024) mlco2\/codecarbon: v2.4.https:\/\/doi.org\/10.5281\/zenodo.11171501","DOI":"10.5281\/zenodo.11171501"},{"key":"1437_CR21","unstructured":"Cruz L (2021) Green software engineering done right: a scientific guide to set up energy efficiency experiments. https:\/\/luiscruz.github.io\/2021\/10\/10\/scientific-guide.html. Accessed 25 Sept 2024"},{"key":"1437_CR22","unstructured":"Cruz L, de\u00a0Bekker P (2023-05-13) All you need to know about energy metrics in software engineering. http:\/\/luiscruz.github.io\/2023\/05\/13\/energy-units.html"},{"key":"1437_CR23","doi-asserted-by":"publisher","unstructured":"Cruz L, Franch X, Mart\u00ednez-Fern\u00e1ndez S (2024) Innovating for tomorrow: the convergence of SE and Green AI. In: SE 2030, p 6. https:\/\/doi.org\/10.1145\/3712007","DOI":"10.1145\/3712007"},{"key":"1437_CR24","doi-asserted-by":"crossref","unstructured":"Deng J, Dong W, Socher R, et\u00a0al (2009) Imagenet: a large-scale hierarchical image database. In: IEEE CVPR. IEEE, pp 248\u2013255","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"1437_CR25","volume":"38","author":"R Desislavov","year":"2023","unstructured":"Desislavov R, Mart\u00ednez-Plumed F, Hern\u00e1ndez-Orallo J (2023) Trends in AI inference energy consumption: beyond the performance-vs-parameter laws of deep learning. Sustain Comput Inform Syst 38:100857","journal-title":"Sustain Comput Inform Syst"},{"key":"1437_CR26","doi-asserted-by":"crossref","unstructured":"Dubhir T, Mishra M, Singhal R (2021) Benchmarking of quantization libraries in popular frameworks. In: IEEE big data. IEEE, pp 3050\u20133055","DOI":"10.1109\/BigData52589.2021.9671500"},{"key":"1437_CR27","doi-asserted-by":"publisher","unstructured":"Duran P, Casta\u00f1o J, G\u00f3mez C et al (2024) GAISSALabel: a tool for energy labeling of ML models. ACM FSE, FSE 2024, pp 622\u2013626. https:\/\/doi.org\/10.1145\/3663529.3663811","DOI":"10.1145\/3663529.3663811"},{"key":"1437_CR28","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-32489-6","volume-title":"Contemporary empirical methods in software engineering","author":"M Felderer","year":"2020","unstructured":"Felderer M, Travassos GH (2020) Contemporary empirical methods in software engineering, vol 1286. Springer, Berlin"},{"key":"1437_CR29","doi-asserted-by":"publisher","unstructured":"Flich J, Medina L, Catal\u00e1n I, et\u00a0al (2022) Efficient inference of image-based neural network models in reconfigurable systems with pruning and quantization. In: IEEE ICIP, pp 2491\u20132495. https:\/\/doi.org\/10.1109\/ICIP46576.2022.9897752","DOI":"10.1109\/ICIP46576.2022.9897752"},{"key":"1437_CR30","unstructured":"Gale T, Elsen E, Hooker S (2019) The state of sparsity in deep neural networks. arXiv:1902.09574"},{"key":"1437_CR31","unstructured":"Gartner I (2024) Gartner identifies the top five strategic technology trends in software engineering for 2024. https:\/\/www.gartner.com\/en\/newsroom\/press-releases\/2024-05-16-gartner-identifies-the-top-five-strategic-technology-trends-in-software-engineering-for-2024. Accessed 15 July 2024"},{"key":"1437_CR32","doi-asserted-by":"publisher","unstructured":"Georgiou S, Kechagia M, Sharma T, et\u00a0al (2022) Green AI: Do deep learning frameworks have different costs? In: ICSE, ICSE\u201922, pp 1082-1094. https:\/\/doi.org\/10.1145\/3510003.3510221","DOI":"10.1145\/3510003.3510221"},{"key":"1437_CR33","doi-asserted-by":"crossref","unstructured":"Goel A, Tung C, Lu YH, et\u00a0al (2020) A survey of methods for low-power deep learning and computer vision. In: IEEE WF-IoT. IEEE, pp 1\u20136","DOI":"10.1109\/WF-IoT48130.2020.9221198"},{"key":"1437_CR34","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.12973434","author":"A Gonz\u00e1lez","year":"2025","unstructured":"Gonz\u00e1lez A, Casta\u00f1o J, Franch X et al (2025) Replication package for \u201cImpact of ML Optimization Tactics on Greener Pre-Trained ML Models\u2019\u2019. Zenodo. https:\/\/doi.org\/10.5281\/zenodo.12973434","journal-title":"Zenodo"},{"key":"1437_CR35","doi-asserted-by":"publisher","first-page":"402","DOI":"10.1016\/j.future.2024.01.033","volume":"155","author":"A Guldner","year":"2024","unstructured":"Guldner A, Bender R, Calero C et al (2024) Development and evaluation of a reference measurement model for assessing the resource and energy efficiency of software products and components-Green Software Measurement Model (GSMM). Futur Gener Comput Syst 155:402\u201341. https:\/\/doi.org\/10.1016\/j.future.2024.01.033","journal-title":"Futur Gener Comput Syst"},{"key":"1437_CR36","doi-asserted-by":"crossref","unstructured":"Hampau RM, Kaptein M, Van\u00a0Emden R, et\u00a0al (2022) An empirical study on the Performance and energy consumption of AI containerization strategies for computer-vision tasks on the edge. In: EASE, pp 50\u201359","DOI":"10.1145\/3530019.3530025"},{"key":"1437_CR37","doi-asserted-by":"crossref","unstructured":"Hanafy WA, Molom-Ochir T, Shenoy R (2021) Design considerations for energy-efficient inference on edge devices. In: ACM e-Energy, pp 302\u2013308","DOI":"10.1145\/3447555.3465326"},{"key":"1437_CR38","unstructured":"Hinton G, Vinyals O, Dean J (2015) Distilling the knowledge in a neural network. arXiv:1503.02531"},{"key":"1437_CR39","unstructured":"House TW (2023) Executive order on the safe, secure, and trustworthy development and use of artificial intelligence|The White House. https:\/\/www.whitehouse.gov\/briefing-room\/presidential-actions\/2023\/10\/30\/executive-order-on-the-safe-secure-and-trustworthy-development-and-use-of-artificial-intelligence\/. Accessed 15 July 2024"},{"key":"1437_CR40","unstructured":"HugGAN C (2022) huggan\/wikiart $$\\cdot$$ Hugging Face. https:\/\/huggingface.co\/datasets\/huggan\/wikiart. Accessed 18 July 2024"},{"key":"1437_CR41","doi-asserted-by":"crossref","unstructured":"J\u00e4rvenp\u00e4\u00e4 H, Lago P, Bogner J, et\u00a0al (2024) A synthesis of green architectural tactics for ML-enabled systems. In: ICSE, pp 130\u2013141","DOI":"10.1145\/3639475.3640111"},{"key":"1437_CR42","doi-asserted-by":"crossref","unstructured":"Jedlitschka A, Ciolkowski M, Pfahl D (2008) Reporting experiments in software engineering. Guide to advanced empirical software engineering, pp 201\u2013228","DOI":"10.1007\/978-1-84800-044-5_8"},{"key":"1437_CR43","unstructured":"Krizhevsky A, Nair V, Hinton G (2009) The CIFAR-10 dataset. http:\/\/www.cs.toronto.edu\/~kriz\/cifar.html"},{"key":"1437_CR44","unstructured":"Lamberti A (2024) Deep learning model optimization methods. https:\/\/neptune.ai\/blog\/deep-learning-model-optimization-methods#:~:text=Model%2Doptimization%20techniques%20aim%20to,enhancing)%20the%20model%27s%20task%20performance"},{"key":"1437_CR45","unstructured":"LeCun Y, Cortes C, J.C.\u00a0Burges C (1998) MNIST, Yann LeCun, Corinna Cortes and Chris Burges. http:\/\/yann.lecun.com\/exdb\/mnist\/. Accessed 06 June 2024"},{"key":"1437_CR46","doi-asserted-by":"publisher","unstructured":"Lee C, Jin J, Kim T, et\u00a0al (2024) OWQ: outlier-aware weight quantization for efficient fine-tuning and inference of large language models. In: Wooldridge MJ, Dy JG, Natarajan S (eds) AAAI. AAAI Press, pp 13355\u20131336. https:\/\/doi.org\/10.1609\/AAAI.V38I12.29237","DOI":"10.1609\/AAAI.V38I12.29237"},{"key":"1437_CR47","unstructured":"Li L, Yin Y, Li S, et\u00a0al (2023) $$\\text{M}^3$$it: a large-scale dataset towards multi-modal multilingual instruction tuning. arXiv:2306.04387"},{"key":"1437_CR48","doi-asserted-by":"crossref","unstructured":"Luccioni S, Jernite Y, Strubell E (2024) Power hungry processing: Watts driving the cost of AI deployment? In: ACM FAccT, pp 85\u201399","DOI":"10.1145\/3630106.3658542"},{"key":"1437_CR49","unstructured":"Marsh J (2023) How many watts does a phone charger use? https:\/\/www.energysage.com\/electricity\/house-watts\/. Accessed 07 June 2024"},{"key":"1437_CR50","doi-asserted-by":"publisher","DOI":"10.1145\/3487043","author":"S Mart\u00ednez-Fern\u00e1ndez","year":"2022","unstructured":"Mart\u00ednez-Fern\u00e1ndez S, Bogner J, Franch X et al (2022) Software engineering for AI-based systems: a survey. ACM TOSEM. https:\/\/doi.org\/10.1145\/3487043","journal-title":"ACM TOSEM"},{"key":"1437_CR51","doi-asserted-by":"publisher","unstructured":"De Martino V, Palomba F (2024) Classification and challenges of non-functional requirements in ML-enabled systems: A systematic literature review. https:\/\/doi.org\/10.1016\/j.infsof.2025.107678","DOI":"10.1016\/j.infsof.2025.107678"},{"key":"1437_CR52","doi-asserted-by":"crossref","unstructured":"Morteza A, Chou RA (2024) Distributed matrix multiplication: download rate, randomness and privacy trade-offs. In: Allerton. IEEE, pp 1\u20137","DOI":"10.1109\/Allerton63246.2024.10735263"},{"key":"1437_CR53","doi-asserted-by":"publisher","DOI":"10.1016\/j.enbuild.2023.113036","volume":"289","author":"A Morteza","year":"2023","unstructured":"Morteza A, Yahyaeian AA, Mirzaeibonehkhater M et al (2023) Deep learning hyperparameter optimization: application to electricity and heat demand prediction for buildings. Energy Build 289:113036","journal-title":"Energy Build"},{"issue":"2","key":"1437_CR54","doi-asserted-by":"publisher","first-page":"280","DOI":"10.1109\/TSUSC.2022.3222631","volume":"8","author":"A Noureddine","year":"2022","unstructured":"Noureddine A, Lodeiro MD, Bru N et al (2022) The impact of green feedback on users\u2019 software usage. IEEE Trans Sustain Comput 8(2):280\u2013292","journal-title":"IEEE Trans Sustain Comput"},{"key":"1437_CR55","unstructured":"Nyanko (2024) nyanko7\/danbooru2023 $$\\cdot$$ Datasets. https:\/\/huggingface.co\/datasets\/nyanko7\/danbooru2023. Accessed 18 July 2024"},{"key":"1437_CR56","unstructured":"Paganini M (2023) PyTorch pruning tutorial. https:\/\/pytorch.org\/tutorials\/intermediate\/pruning_tutorial.html. Accessed 30 April 2024"},{"key":"1437_CR57","doi-asserted-by":"crossref","unstructured":"Rajput S, Sharma T (2024) Benchmarking emerging deep learning quantization methods for energy efficiency. In: ICSA-C. IEEE, pp 238\u2013242","DOI":"10.1109\/ICSA-C63560.2024.00049"},{"key":"1437_CR58","unstructured":"Ralph P (2024) EmpiricalStandards. https:\/\/github.com\/acmsigsoft\/EmpiricalStandards\/tree\/master\/docs\/standards. Accessed 20 July 2024"},{"key":"1437_CR59","doi-asserted-by":"publisher","first-page":"10390","DOI":"10.1016\/j.csi.2024.103906","volume":"92","author":"AD Reguero","year":"2025","unstructured":"Reguero AD, Mart\u00ednez-Fern\u00e1ndez S, Verdecchia R (2025) Energy-efficient neural network training through runtime layer freezing, model quantization, and early stopping. CSI 92:10390. https:\/\/doi.org\/10.1016\/j.csi.2024.103906","journal-title":"CSI"},{"key":"1437_CR60","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1145\/3623402","volume":"14","author":"B Rokh","year":"2023","unstructured":"Rokh B, Azarpeyvand A, Khanteymoori A (2023) A comprehensive survey on model quantization for deep neural networks in image classification. ACM TIST 14:6. https:\/\/doi.org\/10.1145\/3623402","journal-title":"ACM TIST"},{"key":"1437_CR61","unstructured":"Sallou J, Cruz L, Durieux T (2023) Energibridge: empowering software sustainability through cross-platform energy measurement. arXiv:2312.13897"},{"key":"1437_CR62","doi-asserted-by":"publisher","unstructured":"Schubert ME, Langerman D, George AD (2024) Benchmarking inference of transformer-based transcription models with clustering on embedded gpus. IEEE Access pp 1\u20131. https:\/\/doi.org\/10.1109\/ACCESS.2024.3426471","DOI":"10.1109\/ACCESS.2024.3426471"},{"issue":"3\u20134","key":"1437_CR63","doi-asserted-by":"publisher","first-page":"591","DOI":"10.1093\/biomet\/52.3-4.591","volume":"52","author":"SS Shapiro","year":"1965","unstructured":"Shapiro SS, Wilk MB (1965) An analysis of variance test for normality (complete samples). Biometrika 52(3\u20134):591\u2013611","journal-title":"Biometrika"},{"issue":"3","key":"1437_CR64","doi-asserted-by":"publisher","first-page":"32","DOI":"10.1109\/MS.2004.1293070","volume":"21","author":"R van Solingen","year":"2004","unstructured":"van Solingen R (2004) Measuring the ROI of software process improvement. IEEE Softw 21(3):32\u20133. https:\/\/doi.org\/10.1109\/MS.2004.1293070","journal-title":"IEEE Softw"},{"key":"1437_CR65","doi-asserted-by":"publisher","DOI":"10.1002\/9781118162934","volume-title":"Sampling","author":"SK Thompson","year":"2012","unstructured":"Thompson SK (2012) Sampling, vol 755. Wiley, New York"},{"key":"1437_CR66","doi-asserted-by":"crossref","unstructured":"Thrush T, Jiang R, Bartolo M, et\u00a0al (2022) Winoground: probing vision and language models for visio-linguistic compositionality. arXiv:2204.03162","DOI":"10.1109\/CVPR52688.2022.00517"},{"issue":"4","key":"1437_CR67","doi-asserted-by":"publisher","DOI":"10.1002\/widm.1507","volume":"13","author":"R Verdecchia","year":"2023","unstructured":"Verdecchia R, Sallou J, Cruz L (2023) A systematic review of Green AI. Wiley Interdiscip Rev Data Min Knowl Discov 13(4):e1507","journal-title":"Wiley Interdiscip Rev Data Min Knowl Discov"},{"key":"1437_CR68","unstructured":"Wen W (2024) Introduction to torch.compile. https:\/\/pytorch.org\/tutorials\/intermediate\/torch_compile_tutorial.html. Accessed 30 April 2024"},{"key":"1437_CR69","doi-asserted-by":"crossref","unstructured":"Wilcoxon F (1992) Individual comparisons by ranking methods. In: Breakthroughs in statistics: methodology and distribution. Springer, pp 196\u2013202","DOI":"10.1007\/978-1-4612-4380-9_16"},{"key":"1437_CR70","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-29044-2","volume-title":"Experimentation in software engineering","author":"C Wohlin","year":"2012","unstructured":"Wohlin C, Runeson P, H\u00f6st M et al (2012) Experimentation in software engineering. Springer, Berlin"},{"key":"1437_CR71","unstructured":"Xiao H, Rasul K, Vollgraf R (2017) Fashion-MNIST: a novel image dataset for benchmarking machine learning algorithms. arXiv:1708.07747"},{"key":"1437_CR72","doi-asserted-by":"crossref","unstructured":"Xu Y, Mart\u00ednez-Fern\u00e1ndez S, Mart\u00ednez M, et\u00a0al (2023) Energy efficiency of training neural network architectures: an empirical study. In: Proceedings of the 56th annual Hawaii international conference on system sciences, pp 781\u2013790","DOI":"10.24251\/HICSS.2023.098"},{"key":"1437_CR73","unstructured":"Zhu L (2019) pytorch-OpCounter. https:\/\/github.com\/Lyken17\/pytorch-OpCounter\/blob\/master\/benchmark\/README.md. Accessed 29 April 2024"}],"container-title":["Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00607-025-01437-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00607-025-01437-8\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00607-025-01437-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,26]],"date-time":"2025-04-26T17:52:03Z","timestamp":1745689923000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00607-025-01437-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4]]},"references-count":73,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2025,4]]}},"alternative-id":["1437"],"URL":"https:\/\/doi.org\/10.1007\/s00607-025-01437-8","relation":{},"ISSN":["0010-485X","1436-5057"],"issn-type":[{"value":"0010-485X","type":"print"},{"value":"1436-5057","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4]]},"assertion":[{"value":"4 October 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 February 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 April 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}}],"article-number":"103"}}