Dataset Preprocessing
As a demonstration, let us first load the iAWE dataset (which has
already been converted to HDF5 format):
Finding the range of voltage for air conditioner 1
We observe minimum voltage of 0 and maximum of 5140. Clearly, these are due to some fault in data collection. These readings should be removed
Removing readings in the dataset where voltage >260 or voltage <160
Now, observing the voltage variation in the same air conditioner as before.
Filtering data from 13 July to 4 August
Downsampling the dataset to 1 minute
Fill large gaps in appliances with zeros and forward-fill small gaps
Prepend and append zeros
Drop missing samples from mains
Find intersection of mains and appliance datetime indicies