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

New
·

4 Weeks

·

Cohort-based Course

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

Course overview

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

Who is this course for

01

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

02

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

03

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

What you’ll get out of this course

Time series basics

  • terms, basic formalism
  • code representations
  • mathematical foundations
  • exploratory data analysis
  • pre-modelling checks

Diagnose data and learning task accurately

  • endogenous and exogenous data
  • multivariate, multi-instance
  • panel and hierarchical data
  • batch and stream data
  • diagnose your use case

Overview of model types and tasks, best practice advice

  • statistical, ML, deep learning, AI & foundation models
  • when to use what
  • build/adopt/extend trade-offs
  • model assurance, fair evaluation
  • benchmarking models

Model composition, pipelines, AutoML

  • forecasting pipelines
  • feature extraction, feature engineering
  • parameter tuning
  • ensemble models
  • AutoML

Probabilistic forecasting

  • prediction intervals and quantiles
  • distributional predictions
  • probabilistic forecasting metrics
  • evaluation and benchmarking

Hierarchical forecasting, global forecasting

  • hierarchical data
  • composite models
  • reconciliation
  • pooling for multilevel data
  • global forecasting, cross-learning

Cutting edge model types and techniques

  • deep learning
  • torch, pytorch-forecasting
  • foundation models
  • Hugging Face and weight repositories
  • weight handling, fine-tuning

Using sktime - the leading forecasting package

  • easily reusable and deployable code snippets
  • easy to extend for your private projects
  • join the community and stay up to date forever!

This course includes

9 interactive live sessions

Lifetime access to course materials

In-depth lessons

Direct access to instructor

Projects to apply learnings

Guided feedback & reflection

Private community of peers

Course certificate upon completion

Maven Satisfaction Guarantee

This course is backed by Maven’s guarantee. You can receive a full refund within 14 days after the course ends, provided you meet the completion criteria in our refund policy.

Course syllabus

Week 1

Apr 15—Apr 20

    Introduction to the Course and Basic Forecasting with sktime

    • Apr

      15

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

      Tue 4/153:00 PM—4:30 PM (UTC)

    Feature Engineering and Tuning

    • Apr

      17

      Live Session with Robert | Feature Engineering and Tuning

      Thu 4/173:00 PM—4:30 PM (UTC)

Week 2

Apr 21—Apr 27

    Introduction to Probabilistic Forecasting

    • Apr

      22

      Live Session with Franz | Introduction to Probabilistic Forecasting

      Tue 4/223:00 PM—4:30 PM (UTC)

    Advanced Probabilistic Forecasting

    • Apr

      24

      Live Session with Franz | Advanced Probabilistic Forecasting

      Thu 4/243:00 PM—4:30 PM (UTC)

Week 3

Apr 28—May 4

    Panel and Hierarchical Forecasting

    • Apr

      29

      Live Session with Felipe | Panel and Hierarchical Forecasting

      Tue 4/293:00 PM—4:30 PM (UTC)

    Deep Learning for Forecasting

    • May

      1

      Live Session with Felipe | Deep Leaerning for Forecasting

      Thu 5/13:00 PM—4:30 PM (UTC)

Week 4

May 5—May 8

    Spatiotemporal Forecasting and Graph Neural Networks

    • May

      6

      Live Session with Mahdi | Spatiotemporal Forecasting and Graph neural Networks

      Tue 5/63:00 PM—4:30 PM (UTC)

    Foundation Models for Forecasting

    • May

      8

      Live Session with Franz | Foundation Models for Forecasting

      Thu 5/83:00 PM—4:00 PM (UTC)

    Writing your own Forecasters and Contributing to sktime

    • May

      8

      Live Session with Felipe | Writing your own Forecasters and Contributing to sktime

      Thu 5/84:00 PM—5:00 PM (UTC)

Meet your instructor

Franz Király, Ph.D.

Franz Király, Ph.D.

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

Franz Király is director of the German Center for Open Source AI. He has 9 years experience as university professor at top tier universities, and 10 years of experience in industrial data science and AI roles, including 5 years at the Alan Turing institute - all of which included mentoring and teaching responsibilities. Dr Király is also the original founder and a core developer of sktime, the leading package for forecasting and AI with time series.

Robert Kübler, Ph.D.

Robert Kübler, Ph.D.

Senior Staff Data Scientist at ALDI Süd, forecasting for retail

Robert Kübler is a Senior Staff Data Scientist at ALDI Süd and a passionate advocate for machine learning, algorithms, and data-driven problem-solving. He studied Mathematics and completed a Ph.D. in Cryptanalysis.

As a scientific author with over 70 articles on Towards Data Science and 1M+ views, Dr Kübler has a proven track record of making complex topics accessible to diverse audiences.

He has years of experience mentoring data scientists and leading real-world projects—from recommender systems to marketing mix modeling.

Felipe Angelim, MSc

Felipe Angelim, MSc

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

Felipe Angelim is a Tech Lead at Mercado Libre, the largest e-commerce platform in Latin America, a core developer at sktime, and the creator of Prophetverse. With over five years of experience in forecasting for finance and retail, he brings deep expertise in applying advanced techniques to real-world problems. By combining mathematical expertise with programming skills, he develops simple and effective solutions to complex challenges. Felipe studied Engineering at École Centrale de Lyon, France.

Mahdi Torabi Rad, Ph.D.

Mahdi Torabi Rad, Ph.D.

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

Mahdi Torabi Rad is a data scientist, engineer, mentor, and YouTube content creator with over 10 years of experience in developing machine-learning models. Dr Torabi Rad creates videos for his YouTube channel MLBoost to share his knowledge, insights, and passion for AI with a wider audience, with a focus on forecasting, optimization, and modelling of complex physical systems.

A pattern of wavy dots

Join an upcoming cohort

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

Apr-May 2025

$890

Dates

Apr 15—May 8, 2025

Payment Deadline

Apr 13, 2025
Get reimbursed

Course schedule

4-6 hours per week

  • Apr 15 - May 8, 2025

    Tue & Thu, 15:00-16:30 UTC


Frequently Asked Questions

A pattern of wavy dots

Join an upcoming cohort

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

Apr-May 2025

$890

Dates

Apr 15—May 8, 2025

Payment Deadline

Apr 13, 2025
Get reimbursed

$890

4 Weeks