overall_sheet = book.add_sheet('overall')
overall_list_title = ['dataset', 'hidden units', '#week'
] + overall_performance_index
overall_list_row = 0
for c in range(len(overall_list_title)):
overall_sheet.write(0, c, str(overall_list_title[c]))
dataset_list_title = ['activities'] + per_class_performance_index
# Go through all bosch datasets
datasets = ['b1']
for datafile in datasets:
feature_filename = 'feature_' + datafile + '.pkl'
# Looking for processed feature data
if os.path.exists(feature_filename):
feature_file = open(feature_filename, mode='r')
feature_dict = pickle.load(feature_file)
feature = AlFeature()
feature.load_from_dict(feature_dict)
else:
feature = load_casas_from_file(datafile,
normalize=True,
per_sensor=True,
ignore_other=False)
feature_file = open(feature_filename, mode='w')
pickle.dump(feature.export_to_dict(), feature_file, protocol=-1)
feature_file.close()
num_samples = feature.x.shape[0]
train_index = []
test_index = []
x_tensor = theano.shared(np.asarray(feature.x,
dtype=theano.config.floatX),
borrow=True)
assert (type(model) == StackedDenoisingAutoencoder)
x_tensor = theano.shared(np.asarray(x, dtype=theano.config.floatX),
borrow=True)
result = model.classify(x_tensor)
predicted_y = result[0]
confusion_matrix = get_confusion_matrix(num_classes=num_classes,
label=y,
predicted=predicted_y)
return confusion_matrix
if __name__ == '__main__':
# Set current directory to local directory
os.chdir(os.path.dirname(os.path.realpath(__file__)))
# Go through all bosch datasets
datasets = ['b1']
for datafile in datasets:
feature_filename = 'feature_' + datafile + '.pkl'
# Looking for processed feature data
if os.path.exists(feature_filename):
feature_file = open(feature_filename, mode='r')
feature_dict = pickle.load(feature_file)
feature = AlFeature()
feature.load_from_dict(feature_dict)
else:
feature = load_casas_from_file(datafile, datafile + '.translate')
feature_file = open(feature_filename, mode='w')
pickle.dump(feature.export_to_dict(), feature_file, protocol=-1)
feature_file.close()
run_test(feature)
from actlearn.data.AlFeature import AlFeature
from actlearn.utils.event_bar_plot import event_bar_plot
from actlearn.utils.AlResult import AlResult
if __name__ == '__main__':
# Set current directory to local directory
os.chdir(os.path.dirname(os.path.realpath(__file__)))
# Go through all bosch datasets
datasets = ['b1', 'b2', 'b3']
for datafile in datasets:
feature_filename = 'feature_' + datafile + '.pkl'
# Looking for processed feature data
if os.path.exists(feature_filename):
feature_file = open(feature_filename, mode='r')
feature_dict = pickle.load(feature_file)
feature = AlFeature()
feature.load_from_dict(feature_dict)
else:
feature = load_casas_from_file(datafile,
normalize=False,
per_sensor=False)
feature_file = open(feature_filename, mode='w')
pickle.dump(feature.export_to_dict(), feature_file, protocol=-1)
feature_file.close()
num_samples = feature.x.shape[0]
train_index = []
test_index = []
week_array = get_boundary(feature, period='week')
learning_result_fname = 'dt_learning_' + datafile + '.pkl'
learning_result = AlResult(result_name='%s decision tree' % datafile,
data_fname=datafile,
def load_casas_from_file(data_filename, translation_filename=None,
dataset_dir='../datasets/bosch/',
normalize=True, per_sensor=True, ignore_other=False):
"""
Load CASAS Data From File
:param data_filename:
:param translation_filename:
:param dataset_dir:
:param normalize:
:param per_sensor:
:param ignore_other:
:return:
"""
# Initialize AlData Structure
data = AlData()
# Load Translation File
if translation_filename is not None:
data.load_sensor_translation_from_file(dataset_dir + translation_filename)
# Load Data File
data.load_data_from_file(dataset_dir + data_filename)
# Some basic statistical calculations
data.calculate_window_size()
data.calculate_mostly_likely_activity_per_sensor()
# Print out data summary
data.print_data_summary()
# Configure Features
feature = AlFeature()
# Pass Activity and Sensor Info to AlFeature
feature.populate_activity_list(data.activity_info)
feature.populate_sensor_list(data.sensor_info)
# feature.DisableActivity('Other_Activity')
# Add lastEventHour Feature
feature.featureWindowNum = 1
feature.add_feature(AlFeatureSensorCount(normalize=normalize))
feature.add_feature(AlFeatureWindowDuration(normalize=normalize))
feature.add_feature(AlFeatureEventHour(normalize=normalize))
feature.add_feature(AlFeatureEventSensor(per_sensor=per_sensor))
feature.add_feature(AlFeatureLastDominantSensor(per_sensor=per_sensor))
feature.add_feature(AlFeatureEventSecond(normalize=normalize))
feature.add_feature(AlFeatureSensorElapseTime(normalize=normalize))
# Select whether disable other activity or not
if ignore_other:
feature.disable_activity('Other_Activity')
# Print Feature Summary
feature.print_feature_summary()
# Calculate Features
feature.populate_feature_array(data.data)
# Return features data
return feature
os.chdir(os.path.dirname(os.path.realpath(__file__)))
# Loading Log configuration
logging.config.fileConfig('../log/log.cfg')
# Initialize AlData Structure
data = AlData()
# Load Translation File
data.load_sensor_translation_from_file(dataset_dir + 'b1.translate')
# Load Data File
data.load_data_from_file(dataset_dir + 'b1')
# Some basic statistical calculations
data.calculate_window_size()
data.calculate_mostly_likely_activity_per_sensor()
# Print out data summary
data.print_data_summary()
# Configure Features
feature = AlFeature()
# Pass Activity and Sensor Info to AlFeature
feature.populate_activity_list(data.activity_info)
feature.populate_sensor_list(data.sensor_info)
# feature.DisableActivity('Other_Activity')
# Add lastEventHour Feature
feature.featureWindowNum = 1
feature.add_feature(AlFeatureSensorCount(normalize=False))
feature.add_feature(AlFeatureWindowDuration(normalize=False))
feature.add_feature(AlFeatureEventHour(normalize=False))
feature.add_feature(AlFeatureEventSensor(per_sensor=False))
feature.add_feature(AlFeatureLastDominantSensor(per_sensor=False))
feature.add_feature(AlFeatureEventSecond(normalize=False))
feature.add_feature(AlFeatureSensorElapseTime(normalize=False))
# Print Feature Summary
feature.print_feature_summary()