2025-11-07 –, Talk Track 1
Learn how to build accurate retail demand forecasts using MLForecast, an open-source Python library that automates feature engineering and machine learning models, with practical examples for common retail scenarios.
Retail demand forecasting presents unique challenges: seasonal patterns vary across product categories, promotional events create demand spikes, and traditional methods often require extensive manual feature engineering. Many data scientists struggle with the complexity of creating robust features while maintaining forecast accuracy across different product types.
This presentation demonstrates how MLForecast simplifies retail demand forecasting through automated feature engineering and built-in cross-validation. Unlike traditional approaches that require manual lag creation and seasonal decomposition, MLForecast automatically generates relevant features while providing interpretable results.
Through hands-on Python examples with retail datasets, you will learn:
- Automated Features: Understanding MLForecast's automated feature engineering including lag features, rolling statistics, and expanding windows
- Transform Methods: Implementing lag transforms (RollingMean, ExpandingMean, ExponentiallyWeightedMean) for retail demand patterns
- External Variables: Incorporating exogenous variables like promotional calendars, holidays, and external factors into forecasting models
- Model Validation: Cross-validation techniques with time-based splits to validate model performance and prevent data leakage
- Uncertainty Analysis: Generating prediction intervals for uncertainty quantification and visualizing forecasts with confidence bands
No previous knowledge expected
Khuyen Tran transforms how data scientists learn and work. She is the author of Production-Ready Data Science: From Prototyping to Production with Python, a comprehensive guide that helps data professionals bridge the gap between experimentation and deployment.
As founder of CodeCut, she publishes daily Python tips in her newsletter that reach over 10,000 views per month and has built a community of 110,000 LinkedIn followers.
Previously an MLOps Engineer and Senior Data Engineer at Accenture, she built enterprise data solutions for clients worldwide.