Skip to main content Start main content

UBDA Training Course: Jump-start AI Training on Time Series Data

Training Course

  • Date

    09 - 20 Jan 2023

  • Organiser

  • Time

    14:30 - 16:00

  • Venue



You are cordially invited to attend a training course spanning of 2 weeks. This training will introduce and demonstrate to you to the characteristics of time series data, its different analyses, and Virtual Desktop in the UBDA HPC (high performance computing) platform. During the two weeks, the course includes UBDA HPC services, the pre-processing and feature extraction of time series data, the pre-trained models in ONNX and TF Hub, examples of classical models on time series data, introduction of the two forecasting methods (exponential smoothing and ARIMA), comparison of the forecasting performance of these two methods with DL methods such as Prophet and XGBoost and discussion of how to use well-known data sets such as Heng Seng index, and Dow Jones index to conduct forecasting analysis. Participants are expected to have basic knowledge of Python. Please find the arrangements below



Jump-start AI Training on Time Series Data (Poster)



9 Jan 2023 to 20 Jan 2023

Venue   Online via Microsoft Teams

Session activities

  • UBDA HPC services (1st online session: 9 Jan 2023, 14:30 - 16:00)
  • Practical Transfer Learning (2nd online session: 11 Jan 2023, 14:30 - 16:00)
  • Examples of Forecasting (3rd online session: 18 Jan 2023, 14:30 - 16:00)
Consultation Sessions  
  • 13 Jan 2023 (online, 11:00 – 12:00)
  • 20 Jan 2023 (online, 11:00 – 12:00)

Registration / Quota  


PolyU staff and students
100 (first-come-first serve via online registration)

Registration   -

Should you have any enquiries, please contact Dr Vincent Ng ( or Dr Heung Wong (

Your browser is not the latest version. If you continue to browse our website, Some pages may not function properly.

You are recommended to upgrade to a newer version or switch to a different browser. A list of the web browsers that we support can be found here