NILMTK

Imagemap
NILMTK Central Metadata Country Appliance type Prior Disaggregation Model Data Dataset metadata building room ElecMeter Appliance dataset meter devices timeframe Buildings Building MeterGroups Meters Appliances Key Dataset id Building id Meter id Dataframe Time Series Columns Power Active Apparent Reactive Energy Apparent Voltage Current Frequency Power factor Phase Angle Selectable Dataset Converters redd iawe ukdale etc... Datastore csv hdf Disaggregate Combinatorial optimisation FHMM exact Hart 85 Feature detectors Cluster Steady states Preprocessing Apply Clip Stats Drop out rate Good sections Histogram Total Energy Core Data aligned classes Dataset buidings: store: metadata: Building(Hashable) elec:<MeterGroup> metadata: about building only Electric (common methods) ElecMeter(Hashable) Appliances: store: key: metadata: MeterGroup elecmeters or meter groups Appliance(Hashable) metadata: Timeframe start: end: Timeframegroup(list) timeframes Functions plots metrics Infrastructure utils version hashable consts node results exceptions docinherit Demos Buildsys 2014 IAWE GJW etc... Phases Convert Import data Train Disaggregate Compare effectiveness Outstanding questions How is labelled data from training (fing ... How are disaggregation results stored? Any process for layering daily patterns? Any thoughts on consumer output? Any thoughts on consumer input (given no ...
hide
NILMTK User Link
hide
Selectable
hide
Dataset Converters
leaf
redd
leaf
iawe
leaf
ukdale
leaf
etc...
hide
Datastore
leaf
csv
leaf
hdf
hide
Disaggregate
leaf
Combinatorial optimisation
leaf
FHMM exact
leaf
Hart 85
hide
Feature detectors
leaf
Cluster
leaf
Steady states
hide
Preprocessing
leaf
Apply
leaf
Clip
hide
Stats
leaf
Drop out rate
leaf
Good sections
leaf
Histogram
leaf
Total Energy
hide
Core
hide
Data aligned classes
hide
Dataset
leaf
buidings:
leaf
store:
leaf
metadata:
hide
Building(Hashable)
leaf
elec:<MeterGroup>
leaf
metadata: about building only
hide
Electric (common methods)
hide
ElecMeter(Hashable)
leaf
Appliances:
leaf
store:
leaf
key:
leaf
metadata:
hide
MeterGroup
leaf
elecmeters or meter groups
hide
Appliance(Hashable)
leaf
metadata:
hide
Timeframe
leaf
start:
leaf
end:
hide
Timeframegroup(list)
leaf
timeframes
hide
Functions
leaf
plots
leaf
metrics
hide
Infrastructure
leaf
utils
leaf
version
leaf
hashable
leaf
consts
leaf
node
leaf
results
leaf
exceptions
leaf
docinherit
hide
Phases
leaf
Convert
leaf
Import data
leaf
Train
leaf
Disaggregate
leaf
Compare effectiveness
hide
Central Metadata User Link
leaf
Country
leaf
Appliance type
leaf
Prior
leaf
Disaggregation Model
hide
Data
hide
Dataset
hide
metadata
hide
building
leaf
room
leaf
ElecMeter
leaf
Appliance
leaf
dataset
leaf
meter devices
leaf
timeframe
hide
Buildings
hide
Building
hide
MeterGroups
hide
Meters
leaf
Appliances
hide
Key
leaf
Dataset id
leaf
Building id
leaf
Meter id
hide
Dataframe
hide
Time Series
hide
Columns
hide
Power
leaf
Active
leaf
Apparent
leaf
Reactive
hide
Energy
leaf
Apparent
leaf
Voltage
leaf
Current
leaf
Frequency
leaf
Power factor
leaf
Phase Angle
hide
Demos
leaf
Buildsys 2014 User Link
leaf
IAWE
leaf
GJW
leaf
etc...
hide
Outstanding questions
leaf
How is labelled data from training (fingerprints) stored?
leaf
How are disaggregation results stored?
leaf
Any process for layering daily patterns?
leaf
Any thoughts on consumer output?
leaf
Any thoughts on consumer input (given no definitive fingerprint database)