Organization
Location
US
Badges
Activity
Challenge Categories
Challenges Entered
Completed
University of Wollongong
Detecting Energy Flexibility in Buildings
Latest submissions
See All| graded | 300513 | ||
| graded | 300504 | ||
| graded | 300497 |
| Participant | Rating |
|---|---|
yusuf_dogu
|
0 |
ryan_sharp
|
0 |
ibrahim_yucel
|
0 |
| Participant | Rating |
|---|---|
|
ryan_sharp
|
0 |
-
WollongongOrBust Flextrack Challenge 2025View
Flextrack Challenge 2025
Final Round Test Set Released 🧨🚀
About 1 month agoI’m hoping soon! My F5 key needs a break and I’ve been a bit distracted at work waiting for these results.
Edit: it’s been over 24 hours now… May we please get confirmed final results? Or are these the final results?
Important Update
About 2 months ago
How will participants be able to perform feature engineering in phase 2? It sounds like the structure of the data will be pretty different from phase 1 which may cause problems.
📹 Townhall Recording & Q&A with Challenge Organisers | How to use digital twin data to predict demand response capacity
2 months agoThat’s an incorrect usage of that word
📹 Townhall Recording & Q&A with Challenge Organisers | How to use digital twin data to predict demand response capacity
2 months agoThe competition overview clearly addresses the t+1 problem!
“This challenge focuses on identifying and estimating demand response activity. Participants are to develop a machine learning model that back-cast (from historic time-series data from buildings) to:
- determine when demand response events were activated and for how long,
- determine how much energy was increased or decreased (over the event duration), compared with normal consumption, as a result of activating demand response mode.
Participants will use ground truth time-series data with known observed demand response events (identified in the form of demand response flags) to learn site consumption behaviour both (i) when demand response mode is not active and (ii) when demand response mode is activated.”
Backcasting in machine learning is the process of generating or predicting unknown historical data using present information. How exactly are we supposed to interpret the quoted text???
📹 Townhall Recording & Q&A with Challenge Organisers | How to use digital twin data to predict demand response capacity
2 months agoIt mentions backcasting which we know means using t+1,2,3, etc. if necessary
Clarification on Future Data and Back Cast
2 months agoIf that’s the case, you need to wipe the current leaderboard and fix the competition description. Why’d the description say we’re to build models that back-cast if we’re now being told that’s against the rules? What a complete waste of time for many teams. Very frustrating.
Please explain what was meant by the following:
“This challenge focuses on identifying and estimating demand response activity. Participants are to develop a machine learning model that back-cast (from historic time-series data from buildings) to:
- determine when demand response events were activated and for how long,
- determine how much energy was increased or decreased (over the event duration), compared with normal consumption, as a result of activating demand response mode.
Participants will use ground truth time-series data with known observed demand response events (identified in the form of demand response flags) to learn site consumption behaviour both (i) when demand response mode is not active and (ii) when demand response mode is activated.”
📹 Townhall Recording & Q&A with Challenge Organisers | How to use digital twin data to predict demand response capacity
2 months agoHello. In the competition description, you state: “Participants are to develop a machine learning model that back-cast (from historic time-series data from buildings) to:…”
A back-cast is using present/future data to predict past values. You state around the 40-minute mark that this actually isn’t allowed.
Your competition description explicitly says back-cast but your Q&A states the opposite. Could we please get clarification?
💬 Feedback & Suggestions
3 months agoWell, I’ve figured out how to find my submission, but it’s been 19 hours with no update. Just submitted with the message, “Successfully enqueued 1 Job.”
Is this wait time expected? I’m coming from Kaggle, so my expectations might be misaligned with how competitions like this usually work.
💬 Feedback & Suggestions
4 months agoHello! Is there a way to track progress on submissions? I made a submission a couple of hours ago and have no clue if it was successful and is just taking time to be scored or if it failed. The only indication that I even made a submission is my 10 daily submissions going to 9.
🏆 Final Results & Next Steps
About 1 month agoIt appears that the final rankings have undergone several revisions. In light of this, I would like to raise one final question for the competition organizers: why were classification metrics not included in the final ranking criteria?
The competition’s stated objectives are to:
The first objective is inherently a classification problem, while the second is a regression problem. Given this dual focus, it seems logical that both types of metrics should contribute to the final evaluation.
In particular, while the F1 score is generally more informative than the Geometric Mean for assessing classification performance, a balanced combination of both could provide a more comprehensive and fair assessment. I suggest this blended approach be considered for both the competition phase and the final round.