def __init__(self,
training_instances: List[Instance],
n: int,
lda=0.1,
max_iter=50,
verbose=False):
"""
:param training_instances: the instances on which to train
:param n: the number of un-poisoned instances in training_instances
- the size of the original data set
:param lda: lambda - for regularization term
:param max_iter - the maximum number of iterations
:param verbose: if True, the solver will be in verbose mode
"""
Learner.__init__(self)
self.set_training_instances(training_instances)
self.n = n
self.lda = lda # lambda
self.max_iter = max_iter
self.verbose = verbose
self.fvs = None
self.labels = None
self.tau = None
self.w = None
self.b = None
# If true, setup problem on train - only use for learners that use
# the same instance of TRIM
self.redo_problem_on_train = True
self.temp_tuple = None
self.irl_selection = np.full(len(self.training_instances), 1)
def __init__(self, training_instances, verbose=False):
Learner.__init__(self)
self.training_instances = training_instances
self.verbose = verbose
self.w = None
self.mean = None
self.std = None
def __init__(self, params=None, training_instances=None):
Learner.__init__(self)
self.weight_vector = None
self.bias = 0
self.c_delta = 0.5
self.c = 1
if params is not None:
self.set_params(params)
if training_instances is not None:
self.set_training_instances(training_instances)
def __init__(self,
base_model=None,
training_instances=None,
attack_alg=None):
Learner.__init__(self)
self.model = Model(base_model)
self.attack_alg = attack_alg # Type: class
self.adv_params = None
self.attacker = None # Type: Adversary
self.set_training_instances(training_instances)
self.iterations = 5 # int: control the number of rounds directly
def __init__(self, training_instances=None, coef=0.25, params=None):
Learner.__init__(self)
self.weight_vector = None # type: np.array(shape=(1))
self.num_features = 0 # type: int
self.coef = coef # type: float
self.bias = 0 # type: int
if training_instances is not None:
self.set_training_instances(training_instances)
if params is not None:
self.set_params(params)
def __init__(self, training_instances=None, params=None):
Learner.__init__(self)
self.weight_vector = None # type: np.array(shape=1)
self.num_features = 0 # type: int
self.hinge_loss_multiplier = 0.5 # type: float
self.max_feature_deletion = 30 # type: int
self.bias = 0 # type: int
if params is not None:
self.set_params(params)
if training_instances is not None:
self.set_training_instances(training_instances)