Survey on the design of a competition for energy disaggregation algorithms
54 responses
Loading...
Loading responses...
What is your affiliation?
No responses yet for this question.
Are you happy to use your real name (or your company's real name) in the competition?
No responses yet for this question.
Would you like the competition to run continuously or yearly?
No responses yet for this question.
Which "disaggregation tracks" would you be interested in competing?
No responses yet for this question.
Kaggle or custom-built platform?
No responses yet for this question.
Would you be happy to submit an executable program instead of data?
No responses yet for this question.
Training
What if we don't provide any training data? Would you still be able to compete?
No responses yet for this question.
How many instances of each appliance do you require in the training dataset?
No responses yet for this question.
Do you plan to train *only* on data provided by the competition?
No responses yet for this question.
If you might train on additional data...
Are you happy to declare that you have trained on additional data?
No responses yet for this question.
Can you train your algorithm twice?
No responses yet for this question.
Features and data
If we could provide data from any geographical region, which would be most relevant for your algorithm(s)?
No responses yet for this question.
Can your algorithm disaggregate American appliances?
No responses yet for this question.
Which features do you *absolutely require* in the input data?
No responses yet for this question.
Which input features, additional to those you listed above, can your algorithm optionally exploit?
No responses yet for this question.
What metadata about each house do you absolutely require?
No responses yet for this question.
What metadata about each house can you optionally exploit?
No responses yet for this question.
Which features does your algorithm output?
No responses yet for this question.
What's the minimum duration of training data that you require?
No responses yet for this question.
How much "warm up" data does your algorithm require for disaggregation?
No responses yet for this question.
After "warming up", what's the minimum duration of data that your algorithm can disaggregate per house?
No responses yet for this question.
Is this sufficient data: 1 year of data across 5 homes for training & 1 year of data across 5 different homes for testing; all data from the US
No responses yet for this question.
Disaggregation categories & use cases
Which NILM use-cases are most relevant to your work?
No responses yet for this question.
Performance metrics
No responses yet for this question.
For which categories does your algorithm estimate energy consumption for?
No responses yet for this question.
Please list which appliances your algorithm can detect
No responses yet for this question.
Which non-appliance categories does your algorithm detect?
No responses yet for this question.
Sample rates
For the input data, what is the fastest sample rate that your algorithm can handle?
No responses yet for this question.
For the input data, what is the slowest sample rate that your algorithm can handle?
No responses yet for this question.
We intend to provide data at five temporal resolutions: 1 Hz; every 10 seconds; minutely; half-hourly and hourly. Are you happy with this set of temporal resolutions?
No responses yet for this question.
If the competition provided only a *single* sample rate then which would you prefer?
No responses yet for this question.
What is the temporal resolution of your algorithm's output?
No responses yet for this question.
Final section :)
Should the competition try to measure the time it takes each algorithm to perform disaggregation?
No responses yet for this question.
Assuming we meet all your requirements, would you be happy to participate in a NILM competition?
No responses yet for this question.
Any other suggestions?
No responses yet for this question.
Number of daily responses
This content is neither created nor endorsed by Google. Report Abuse - Terms of Service - Additional Terms