In [1]:

    

%matplotlib inline
from __future__ import print_function, division
from neuralnilm.source import RandomSegments, RandomSegmentsInMemory, SameLocation
import matplotlib.pyplot as plt
import matplotlib as mpl
mpl.rcParams['figure.figsize'] = (10.0, 8.0)
import numpy as np
from sys import stdout


    

In [2]:

    

source = SameLocation(
    filename='/data/mine/vadeec/merged/ukdale.h5',
    #target_appliance=['washer dryer', 'washing machine'],
    target_appliance=['fridge freezer', 'fridge'],
    #window=("2013-03-18", None),
    window=("2013-03-18", "2013-05-18"),
    #window=("2013-04-01", "2013-08-01"),
    seq_length=512,
    train_buildings=[1],#,2,3,4,5],
    validation_buildings=[1],#,2,3,4,5],
    n_seq_per_batch=4,
    standardise_input=True,
    independently_center_inputs=True,
    standardise_targets=True,
    subsample_target=8,
    allow_incomplete=True,
    ignore_incomplete=True,
    include_all=True,
    skip_probability=0.25,
    offset_probability=1
    #ignore_offset_activations=True
)


    

    




Loaded 1508 fridge freezer activations from house 1.






    




INFO:neuralnilm.source:Loaded 1508 fridge freezer activations from house 1.






    




Loaded mains data for building 1.






    




INFO:neuralnilm.source:Loaded mains data for building 1.

In [3]:

    

source.validation_buildings


    

    
Out[3]:





[1]

In [4]:

    

source.train_buildings


    

    
Out[4]:





[1]

In [5]:

    

def plot_data(X, y, seq_i=0):
    fig, axes = plt.subplots(nrows=2, sharex=False)
    axes[0].plot(y[seq_i, :, 0])
    axes[1].plot(X[seq_i, :, 0])
    plt.show()


    

In [6]:

    

source.activations[1][13].plot()


    

    
Out[6]:





<matplotlib.axes._subplots.AxesSubplot at 0x7f7a60cc97d0>






    

In [7]:

    

len(source.activations[1])


    

    
Out[7]:





1508

In [20]:

    

#source.n_seq_per_batch = 4
#source.skip_probability = 0.5
source.ignore_incomplete = False
X_val, y_val = source.validation_data()


    

In [21]:

    

plot_data(X_val, y_val, 0)


    

In [22]:

    

source.start()


    

In [135]:

    

source.ignore_incomplete = True
X, y = source.queue.get(timeout=5)


    

In [136]:

    

plot_data(X, y, 0)


    

In [18]:

    

i = 0
while True:
    print("\r", i, end="")
    stdout.flush()
    X, y = source.queue.get(timeout=5)
    if (X[0] == X_val[0]).all():
        print("ahhh!")
        break
    i += 1


    

    




 168629





    




---------------------------------------------------------------------------
KeyboardInterrupt                         Traceback (most recent call last)
<ipython-input-18-8a1a8d697bca> in <module>()
      3     print("\r", i, end="")
      4     stdout.flush()
----> 5     X, y = source.queue.get(timeout=5)
      6     if (X[0] == X_val[0]).all():
      7         print("ahhh!")

/usr/lib/python2.7/Queue.pyc in get(self, block, timeout)
    175                     if remaining <= 0.0:
    176                         raise Empty
--> 177                     self.not_empty.wait(remaining)
    178             item = self._get()
    179             self.not_full.notify()

/usr/lib/python2.7/threading.pyc in wait(self, timeout)
    357                         break
    358                     delay = min(delay * 2, remaining, .05)
--> 359                     _sleep(delay)
    360                 if not gotit:
    361                     if __debug__:

KeyboardInterrupt: 

In [6]:

    

source.dataset.buildings[1].elec['washer dryer'].dominant_appliance().type


    

    
Out[6]:





{'categories': {'electrical': ['single-phase induction motor', 'resistive'],
  'google_shopping': ['laundry appliances'],
  'size': 'large',
  'traditional': 'wet'},
 'components': [{'categories': {'electrical': ['single-phase induction motor']},
   'description': 'drum motor',
   'n_ancestors': 1,
   'parent': 'component',
   'type': 'motor'},
  {'categories': {'electrical': ['single-phase induction motor']},
   'n_ancestors': 2,
   'parent': 'motor',
   'type': 'water pump'},
  {'additional_properties': {'volume': {'description': 'litres',
     'type': 'number'}},
   'categories': {'electrical': ['resistive']},
   'n_ancestors': 2,
   'parent': 'electric heating element',
   'type': 'electric water heater'},
  {'categories': {'electrical': ['resistive']},
   'description': 'clothes drying heating element',
   'n_ancestors': 2,
   'parent': 'electric heating element',
   'type': 'electric air heater'}],
 'control': ['manual', 'timer'],
 'distributions': {'rooms': [{'distance': 3,
    'distribution_of_data': {'categories': ['kitchen',
      'utility',
      'basement',
      'garage'],
     'values': [0.3, 0.3, 0.2, 0.1]},
    'from_appliance_type': 'wet appliance',
    'source': 'subjective'}]},
 'min_off_duration': 600,
 'min_on_duration': 1800,
 'n_ancestors': 4,
 'parent': 'washing machine',
 'subtypes': ['top-loader', 'front-loader'],
 'type': 'washer dryer'}

In [9]:

    

min([a.index[-1] - a.index[0] for a in source.activations[1]])


    

    
Out[9]:





Timedelta('0 days 00:07:42')

In [46]:

    

(2*60*60) / 6


    

    
Out[46]:





1200.0

In [3]:

    

elec = source.dataset.buildings[4].elec


    

In [4]:

    

washer = elec['washing machine']


    

In [5]:

    

data = washer.power_series_all_data()


    

In [6]:

    

data.plot()


    

    
Out[6]:





<matplotlib.axes._subplots.AxesSubplot at 0x7f56fdc044d0>

In [ ]: