Forecasting with Sktime – Introduction and Advanced Features: Pipelines, Hierarchical and Probabilistic Forecasts, Deep Learning and Foundation Models

Abstract: 

sktime is the most widely used scikit-learn compatible framework library for learning with time series. sktime is maintained by a neutral non-profit under permissive license, easily extensible by anyone, and interoperable with the python data science stack.

This workshop gives a hands-on introduction to forecasting with sktime, including advanced features such as hierarchical and probabilistic forecasts , and an overview of different model categories, including classical statistical, ML, deep learning, and foundation models.

Session Outline:

The workshop will provide a hands-on introduction to basic forecasting with sktime and advanced topics:

1. introduction - common forecasting use cases; endogenous and exogenous data, horizons, basic vignette; simple statistical models, reduction to sklearn regressors; navigating the model marketplace

2. forecasting pipelines - feature extraction, transformations for endogenous and exogenous data

3. tuning, evaluation and automl - parameter fitters, seasonality/stationarity, backtesting based tuning, multiplexing and automl

4. advanced features: probabilistic and hierarchical forecasting; probabilistic metrics;

5. advanced model classes: global forecasters, deep learning based forecasters, foundation models

Background Knowledge:

At least beginner knowledge in python is expected. Notebooks can be run in the cloud, but ability to set up environments locally is recommended.

Bio: 

Marc Rovira is a data scientist at Electrolux Group in Stockholm, with a strong focus on forecasting and time series analysis. He actively contributes to the sktime community as a council member and user representative. Prior to his industry experience, Marc completed a Ph.D. that explored the intersection of computational fluid mechanics, chemical engineering, and machine learning, with the aim of mitigating air pollution. His educational background also includes a master's degree in aerospace engineering.

Open Data Science

 

 

 

Open Data Science
One Broadway
Cambridge, MA 02142
info@odsc.com

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Youtube
Consent to display content from - Youtube
Vimeo
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google