def training(features, target, cycle):
try:
if cycle != 1:
model = repository.get(PassiveAggressiveClassifier.__name__,
DIR_PATH)
scaler = repository.get(StandardScaler.__name__, DIR_PATH)
else:
model = PassiveAggressiveClassifier(loss='squared_hinge')
scaler = StandardScaler()
# add the new data to the existing scaler model, because the scaling depends on the data
scaler.partial_fit(features)
# scale the new features
features = scaler.transform(features)
# Perform online learning with the new features
model.partial_fit(features, target, classes=np.array([0, 1]))
# Store the model and scaler to respective files for later use
repository.create(model, PassiveAggressiveClassifier.__name__,
DIR_PATH)
repository.create(scaler, StandardScaler.__name__, DIR_PATH)
return {'message': 'training successful'}
except Exception as e:
traceback.print_tb(e.__traceback__)
return {'message': 'training failed '}
def training(features, target):
try:
model = repository.get(SGDClassifier(loss='hinge', penalty='l1'),
SGDClassifier.__name__, DIR_PATH)
scaler = repository.get(MinMaxScaler(), MinMaxScaler.__name__,
DIR_PATH)
scaler.partial_fit(features)
features = scaler.transform(features)
model.partial_fit(features, target, classes=np.array([0, 1]))
repository.create(model, SGDClassifier.__name__, DIR_PATH)
repository.create(scaler, MinMaxScaler.__name__, DIR_PATH)
return {'message': 'training successful'}
except Exception as e:
traceback.print_tb(e.__traceback__)
return {'message': 'training failed '}
def training(features, target):
try:
model = repository.get(
SGDRegressor(loss='epsilon_insensitive', penalty='l2'),
SGDRegressor.__name__, DIR_PATH)
scaler = repository.get(MaxAbsScaler(), MaxAbsScaler.__name__,
DIR_PATH)
scaler.partial_fit(features)
features = scaler.transform(features)
model.partial_fit(features, target)
repository.create(model, SGDRegressor.__name__, DIR_PATH)
repository.create(scaler, MaxAbsScaler.__name__, DIR_PATH)
return {'message': 'training successful '}
except Exception as e:
traceback.print_tb(e.__traceback__)
return {'message': 'training failed '}
def training(features, target, cycle):
try:
if cycle != 1:
model = repository.get(SGDClassifier.__name__, DIR_PATH)
scaler = repository.get(MinMaxScaler.__name__, DIR_PATH)
else:
model = SGDClassifier(loss='hinge', penalty='l1')
scaler = MinMaxScaler()
scaler.partial_fit(features)
features = scaler.transform(features)
# online learn the adaptation options (features) and their targets (goal satisfactions)
model.partial_fit(features, target, classes=np.array([0, 1, 2, 3]))
# Save the scaler and classifier model
repository.create(model, SGDClassifier.__name__, DIR_PATH)
repository.create(scaler, MinMaxScaler.__name__, DIR_PATH)
return {'message': 'training successful'}
except Exception as e:
traceback.print_tb(e.__traceback__)
return {'message': 'training failed '}