A python package for time series forecasting with scikit-learn estimators.

flexible time series forecasting using machine learning

Portfolio project: Machine learning automation project for a large American retailer. Reduced warehouse costs and stock-outs by developing a scalable set of recursive machine learning models that predict the demand in the next 8 days at store-product level based on the data contained in the three-year-history company’s SQL database.