Automating Machine Learning Model Development: An OperationalML Approach with PyCARET and Streamlit

OperationalML (Automatic Machine Learning) refers to tools and techniques that automate various machine-learning model development aspects, including data preparation, feature engineering, model selection, hyperparameter optimization, and evaluation. OperationalML can significantly reduce the time and expertise required for building high-quality machine- learning models while making the process more transparent and scalable. In this study, we present an implementation of OperationalML using two open-source libraries: PyCARET and Streamlit. Our work focuses on evaluating the performance and usability of our implementation across classification tasks from publicly available datasets. Results show that our OperationalML system outperforms manual approaches, especially in speed and ease of use. Additionally, we conduct sensitivity analysis to assess the effectiveness of each component within the system and offer insights into future improvements that could enhance the overall quality of the generated models. This work contributes to the growing body of literature on Machine learning in Production, serving as a valuable resource for practitioners, researchers, and educators who want to quickly develop effective predictive models without spending extensive time on training machine learning.