Deploying Machine Learning Models with Python: Balancing Speed, Accuracy, and Sustainability
DOI:
https://doi.org/10.15680/IJCTECE.2022.0506002Keywords:
Machine Learning Deployment, Python Frameworks, Model Optimization, Inference Speed, Model Accuracy, Sustainable AI, Containerization (Docker, Kubernetes), Cloud and Edge DeploymentAbstract
Deploying machine learning (ML) models effectively requires balancing three critical factors: speed, accuracy, and sustainability. This paper explores strategies and tools within the Python ecosystem that facilitate the deployment of ML models while optimizing for these factors. We discuss model optimization techniques, deployment frameworks, and sustainability considerations, providing a comprehensive guide for practitioners aiming to deploy efficient and eco-friendly ML solutions.
References
1. Matthew Mayo. "Tips for Deploying Machine Learning Models Efficiently." Machine Learning Mastery. https://machinelearningmastery.com/tips-deploying-machine-learning-models-efficiently/MachineLearningMastery.com
2. Toxigon. "Deploying Machine Learning Models with Python: A Step-by-Step Guide." Toxigon Blog. https://toxigon.com/deploying-machine-learning-models-with-pythonToxigon
3. Meghana Tedla, Shubham Kulkarni, Karthik Vaidhyanathan. "EcoMLS: A Self-Adaptation Approach for Architecting Green ML-Enabled Systems." arXiv. https://arxiv.org/abs/2404.11411arXiv
4. Bytescrum. "How to Deploy Machine Learning Models in Production: Key Challenges and Fixes." Bytescrum Blog. https://blog.bytescrum.com/how-to-deploy-machine-learning-models-in-productionBytescrum Blog
5. Crest Infotech. "Deploying Machine Learning Models with Python: Best Practices and Tools." Crest Infotech. https://www.crestinfotech.com/deploying-machine-learning-models-with-python-best-practices-and-tools/Crest Infotech
