Welcome to the workshop!
Workshop
In ProgressWelcome to Advanced Feature Engineering for Time-Series Data!
Today's icebreaker:
The worst prediction ever!
Predictions about the future don’t always go as planned. Can you think of a personal, historical or pop culture prediction that was completely wrong?
Share your best response in the chat!
Today's agenda:
✨
Review:
Recap key concepts.
✨
Practise:
Time-series features for energy prediction.
✨
Closing:
Key takeaways and next steps.
Today's learning objectives:
✨
Create time-based features from raw timestamp data
to capture temporal patterns and improve predictive model performance.
✨
Incorporate domain-specific contextual data
into feature engineering to enhance model relevance and accuracy.
✨
Apply feature selection or dimensionality reduction techniques
to optimise model performance when working with a large number of engineered features.
✨
Explain how engineered features would be maintained and monitored
in a production environment to ensure continued model performance.
Workshop slides
Preview or download a copy of the workshop 2 slides:
PDF preview loads in your browser…
If it doesn’t appear, download it here.
In progress