Facilitating Model Training with Automated Techniques
Abstract
Automating artificial intelligence (AI) model training has emerged as a significant challenge in the field of automation. The complete pipeline from raw data to model deployment poses the need to define robust processes that ensure the efficiency of the services that expose the models. This paper introduces a generic architecture for automating data preparation, training of models, selection of models, and deployment of models as web services for third-party consumption using Microsoft Azure Machine Learning’s (AzureML) CI/CD tools. We conducted a practical experiment utilizing AzureML pipelines with predefined and custom modules, demonstrating its readiness for integration into any production application. We also successfully integrated this architecture into a real-world product designed for industrial forecasting. This practical implementation demonstrates the effectiveness and adaptability of our approach, indicating its potential to address diverse training needs.
References
[2] Swagger: The world’s most popular framework for apis. https://swagger.io, 2022. Accessed: Feb. 7, 2022.
[3] Algorithmia. 2021 state pf enterprise ml. https://info.algorithmia.com/hubfs/2020/Reports/2021-Trends-in-ML/Algorithmia_2021_enterprise_ML_trends.pdf. Accessed: June 9, 2023.
[4] Ashmore, R., Calinescu, R., and Paterson, C. Assuring the machine learning lifecycle: Desiderata, methods, and challenges. ACM Computing Surveys (CSUR) 54, 5 (2021), 1–39.
[5] Boer, A., Koolen, M., van den Berg, J., and van der Werf, J. Continuous delivery pipelines: Best practices in safe. In 2018 IEEE 15th International Conference on e-Business Engineering (ICEBE) (2018), IEEE, pp. 217–224.
[6] Chung, J., Gulcehre, C., Cho, K., and Bengio, Y. Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555 (2014).
[7] Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018).
[8] Feretzakis, G., Sakagianni, A., Loupelis, E., Kalles, D., Skarmoutsou, N., Martsoukou, M., Christopoulos, C., Lada, M., Petropoulou, S., Velentza, A., et al. Machine learning for antibiotic resistance prediction: A prototype using off-the-shelf techniques and entry-level data to guide empiric antimicrobial therapy. Healthcare informatics research 27, 3 (2021), 214–221.
[9] Fischer, D., Jung, M., Kasparick, M., and Momm, C. Continuous integration and deployment for software of things using jenkins and docker: A case study. In 2019 IEEE International Conference on Software Architecture Companion (ICSA-C) (2019), IEEE, pp. 9–16.
[10] Hochreiter, S., and Schmidhuber, J. Long short-term memory. Neural computation 9, 8 (1997), 1735–1780.
[11] Huang, W. Best practices for continuous integration and delivery. In 2016 11th International Conference on Computer Science & Education (ICCSE) (2016), IEEE, pp. 422–426.
[12] Hutter, F., Kotthoff, L., and Vanschoren, J. Automated machine learning: Methods, systems, challenges. arXiv preprint arXiv:1908.02259 (2019).
[13] Kim, M., Zimmermann, T., DeLine, R., and Begel, A. Data scientists in software teams: State of the art and challenges. IEEE Transactions on Software Engineering 44, 11 (2017), 1024–1038.
[14] Li, L., Jamieson, K., DeSalvo, G., Rostamizadeh, A., and Talwalkar, A. Automating neural architecture search using bayesian optimization and hyperband. In International Conference on Learning Representations (2018).
[15] Li, W., Chen, J., and Huang, W. A survey on continuous integration, delivery and deployment tools. Journal of Systems and Software 147 (2018), 1–15.
[16] Martin, R. C. The single responsibility principle. The principles, patterns, and practices of Agile Software Development (2002), 149–154.
[17] Merkel, D. Docker: lightweight linux containers for consistent development and deployment. Linux journal 2014, 239 (2014), 2.
[18] Microsoft. Azure machine learning. https://azure.microsoft.com/en-us/services/machine-learning/, 2021. Accessed: February 7, 2023.
[19] Microsoft. Microsoft customer stories. Microsoft Azure Blog (January 2022).
[20] Paleyes, A., Urma, R.-G., and Lawrence, N. D. Challenges in deploying machine learning: a survey of case studies. ACM Computing Surveys 55, 6 (2022), 1–29.
[21] Pliuhin, V., Pan, M., Yesina, V., and Sukhonos, M. Using azure maching learning cloud technology for electric machines optimization. In 2018 International Scientific-Practical Conference Problems of Infocommunications. Science and Technology (PIC S&T) (2018), IEEE, pp. 55–58.
[22] Rajagopal, S., Hareesha, K. S., and Kundapur, P. P. Performance analysis of binary and multiclass models using azure machine learning. International Journal of Electrical & Computer Engineering (2088-8708) 10, 1 (2020).
[23] Rezazadeh, A. A generalized flow for b2b sales predictive modeling: An azure machine-learning approach. Forecasting 2, 3 (2020), 267–283.
[24] Sculley, D., Holt, G., Golovin, D., Davydov, E., Phillips, T., Ebner, D., Chaudhary, V., Young, M., Crespo, J.-F., and Dennison, D. Hidden technical debt in machine learning systems. Advances in neural information processing systems 28 (2015).
[25] Sharir, O., Peleg, B., and Shoham, Y. The cost of training nlp models: A concise overview. arXiv preprint arXiv:2004.08900 (2020).
[26] Tan, M., and Le, Q. V. Efficientnet: Rethinking model scaling for convolutional neural networks. In International Conference on Machine Learning (2020), PMLR, pp. 6105–6114.
[27] Umamahesan, A., and Babu, D. M. I. From zero to ai hero with automated machine learning. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (2020), pp. 3495–3495.
[28] Wagstaff, K. L., Doran, G., Davies, A., Anwar, S., Chakraborty, S., Cameron, M., Daubar, I., and Phillips, C. Enabling onboard detection of events of scientific interest for the europa clipper spacecraft. In Proceedings of the 25th acm sigkdd international conference on knowledge discovery & data mining (2019), pp. 2191–2201
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
When the article is accepted for publication, I, as the author and representative of the coauthors, hereby agree to transfer to Studia Universitatis Babes-Bolyai, Series Informatica, all rights, including those pertaining to electronic forms and transmissions, under existing copyright laws, except for the following, which the author specifically retain: the right to make further copies of all or part of the published article for my use in classroom teaching; the right to reuse all or part of this material in a review or in a textbook of which I am the author; the right to make copies of the published work for internal distribution within the institution that employs me.
