Smart Approach in Development and Deployment Process for ML Models

Learn how to optimize machine learning model development through smart data preparation strategies in this 24-minute conference talk from Data Science Conference Europe 2022. Discover how implementing a smart generic data mart can significantly reduce production time for new ML models, with approximately 80% of development time typically spent on data preparation. Explore practical solutions for handling data transfer challenges between Data Warehouses and Data Lakes, along with effective approaches to data preprocessing, development, and deployment/orchestration of ML models using Python/PySpark scripts. Gain insights into managing both structured and unstructured data sources while streamlining the overall machine learning development process.