Graham Wood
Graham Wood
Probably not the same thing, ignoring for a minute constantly cycling appliances like fridges, the fingerprint for a washing machine might be: - result label: 'Washing Machine1' - metadata: as...
@JackKelly @nipunreddevil (as you are looking at Hart85 soon) After having explored the implementation of Hart in nilmtk at some length, and having read Hart's '85 and '92 papers, I...
While I remember - For devices with sub components in the metadata, the list of centroids could be derived by a tree walk of the sub components if the centroids...
Ok, I will stumble around a bit, and document what I do I can't seem to find my appliances in the metadata, although they were in the yaml file and...
Slightly confused by "dominant_appliance" as I don't have any defined (as I have no sub-metering). I get the following ``` python In [22]: house = elec['fridge'] #only one meter so...
Got it! ``` python In [27]: house.appliances[1].metadata house.appliances[1].metadata Out[27]: {'additional_properties': {'fuel': 'natural gas'}, 'brand': 'Solo powered flue', 'building': 1, 'dataset': 'gjwdata', 'dates_active': None, 'description': 'Gas fired central heating boiler', 'instance':...
Starter for 10 below with some basic signature generation # Metadata Experiments ## Imports ``` python In [1]: from __future__ import print_function, division from nilmtk import DataSet from nilm_metadata import...
I think this shows it better, I set the threshold to 1000, to get only large devices, it looks like the reactive value of the centroid is the absolute rather...
@nipunbatra @JackKelly If you guys have finished your theses now please could you review the various Hart questions (including this one) that I raised in October. If not let me...
I saw the post from @oliparson about the workshop, I won't be going but would be interested in what the speakers have to say, particularly George Hart. (Also your guest...