Dataset adapters

The general idea behind dataset adapters is to abstract loading of datasets into the general appliance framework. This notebook will show some common ways of using them.



In [1]:

    
import sys
sys.path.append('../../')

import disaggregator as dis

An example of how to initialize a database adapter pecan street.



In [2]:

    
db_url = "postgresql://USERNAME:PASSWORD@db.wiki-energy.org:5432/postgres"
ps_adapter = dis.PecanStreetDatasetAdapter(db_url)

Grab all unique dataids for a particular month



In [3]:

    
dataids = ps_adapter.get_unique_dataids("curated",1,2013,1)



In [4]:

    
dataids[0]









    Out[4]:





5949

Get month-long traces for a home



In [5]:

    
traces = ps_adapter.get_month_traces("curated",2013,1,5949,group=1,sampling_rate="15T")



In [6]:

    
traces[0]









    Out[6]:





<disaggregator.ApplianceTrace.ApplianceTrace at 0x112086110>



In [10]:

    
traces[0].get_series().name









    Out[10]:





u'air1'



In [ ]: