Python Forecasting Masterclass: From Foundations to State-of-the-Art

Robert Kübler, Ph.D.

senior data scientist retail forecasting

Franz Király, Ph.D.

creator sktime forecasting toolbox

+ Mahdi Torabi Rad, Ph.D. and Felipe Angelim, MSc

Master forecasting in Python and join a growing community of practice

- From Python forecasting basics to foundation models in four weeks

- Use-case driven with examples, methodologically solid

- Learn to code in sktime, the leading forecasting framework

- Onboard to the sktime community of practice - stay up-to-date forever!

- No biased product placement or hidden advertisement of paid services

- Become an expert and support open source - 50% towards sktime maintenance and development

What you’ll learn

A gentle introduction for newcomers, and a deep-dive into advanced topics such as hierarchical forecasting, foundation models

  • endogenous and exogenous data

  • multivariate, multi-instance

  • panel and hierarchical data

  • forecasting pipelines, feature extraction

  • parameter tuning, AutoML

  • ensemble models

  • prediction intervals and quantiles, distributional forecasts

  • probabilistic forecasting metrics

  • evaluation and benchmarking

  • hierarchical composite models, pooling

  • hierarchical reconciliation

  • global forecasting, cross-learning

  • deep learning, torch, pytorch-forecasting

  • foundation models, weight handling, pre-training & fine-tuning

  • Hugging Face and weight repositories

  • easily reusable and deployable code snippets

  • easy to extend for your private projects

  • join the community and stay up to date forever!

Learn directly from expert instructors

Robert Kübler, Ph.D.

Robert Kübler, Ph.D.

Senior Staff Data Scientist at ALDI Süd, forecasting for retail

Franz Király, Ph.D.

Franz Király, Ph.D.

Director, German Center of Open Source AI, and founder of sktime

Mahdi Torabi Rad, Ph.D.

Mahdi Torabi Rad, Ph.D.

Content creator tech & AI, founder and data science consultant and MLBoost

Felipe Angelim, MSc

Felipe Angelim, MSc

sktime core developer, forecasting for e-commerce, finance, and retail

Who this course is for

  • Ideal for beginners with a grasp of machine learning. Learn time series forecasting in Python, from the ground up.

  • Expand and complete your toolkit, stay up-to-date. Learn about the newest techniques and cutting-edge methods.

  • For professional practitioners. Learn from basics to advanced techniques, tailored for real-world impact, with use-case examples.

What's included

Live sessions

Learn directly from your instructors in a real-time, interactive format.

Lifetime access

Go back to course content and recordings whenever you need to.

Community of peers

Stay accountable and share insights with like-minded professionals.

Certificate of completion

Share your new skills with your employer or on LinkedIn.

Maven Guarantee

This course is backed by the Maven Guarantee. Students are eligible for a full refund up until the halfway point of the course.

Course syllabus

Week 1

Nov 24—Nov 30

    Introduction to the Course and Basic Forecasting with sktime

    • Nov

      24

      Live Session with Robert | Introduction to the Course and Basic Forecasting with sktime

      Mon 11/244:00 PM—5:30 PM (UTC)

    Feature Engineering and Tuning

    • Nov

      26

      Live Session with Robert | Feature Engineering and Tuning

      Wed 11/264:00 PM—5:30 PM (UTC)

Week 2

Dec 1—Dec 7

    Panel and Hierarchical Forecasting

    • Dec

      1

      Live Session with Felipe | Panel and Hierarchical Forecasting

      Mon 12/14:00 PM—6:00 PM (UTC)

    Deep Learning for Forecasting

    • Dec

      3

      Live Session with Felipe | Deep Learning for Forecasting

      Wed 12/34:00 PM—5:30 PM (UTC)

Free resource

AI for forecasting: foundations models, hype & best practice cover image

AI for forecasting: foundations models, hype & best practice

forecasting foundation models - what they are, what they do

Gentle introduction to foundation and pre-trained models. Zero-shot, few-shot, fine-tuning; open vs gated, etc

best practice - handling software, weights, deps, evaluation

special considerations and "FMops" for forecasting - FM handling (best) practice, key difficulties & solutions

... and what the vendors don't want you to know about

lock-in tactics, influencer propaganda, gated/closed models, licenses, full-value open alternatives

Schedule

Live sessions

5 hrs / week

    • Mon, Nov 24

      4:00 PM—5:30 PM (UTC)

    • Wed, Nov 26

      4:00 PM—5:30 PM (UTC)

    • Mon, Dec 1

      4:00 PM—6:00 PM (UTC)

Projects

2 hrs / week

Async content

1 hr / week

Frequently asked questions