def __init__(self, model, max_iter=1000, C=1.0, verbose=0, n_jobs=1,
show_loss_every=0, logger=None, batch_mode=False,
line_search=True, check_dual_every=10, tol=.001,
do_averaging=True):
if n_jobs != 1:
raise ValueError("FrankWolfeSSVM does not support multiprocessing"
" yet. Ignoring n_jobs != 1.")
BaseSSVM.__init__(self, model, max_iter, C, verbose=verbose,
n_jobs=n_jobs, show_loss_every=show_loss_every,
logger=logger)
self.tol = tol
self.batch_mode = batch_mode
self.line_search = line_search
self.check_dual_every = check_dual_every
self.do_averaging = do_averaging
def __init__(self,
model,
max_iter=1000,
C=1.0,
verbose=0,
n_jobs=1,
show_loss_every=0,
logger=None,
batch_mode=False,
line_search=True,
check_dual_every=10,
tol=.001,
do_averaging=True,
sample_method='perm',
random_state=None,
X_test=None,
Y_test=None):
if n_jobs != 1:
warnings.warn("FrankWolfeSSVM does not support multiprocessing"
" yet. Ignoring n_jobs != 1.")
if sample_method not in ['perm', 'rnd', 'seq']:
raise ValueError("sample_method can only be perm, rnd, or seq")
BaseSSVM.__init__(self,
model,
max_iter,
C,
verbose=verbose,
n_jobs=n_jobs,
show_loss_every=show_loss_every,
logger=logger)
self.tol = tol
self.batch_mode = batch_mode
self.line_search = line_search
self.check_dual_every = check_dual_every
self.do_averaging = do_averaging
self.sample_method = sample_method
self.random_state = random_state
self.X_test = X_test
self.Y_test = Y_test
self.oracle_errs = []
def __init__(self,
model,
max_iter=10000,
C=1.0,
check_constraints=False,
verbose=0,
negativity_constraint=None,
n_jobs=1,
break_on_bad=False,
show_loss_every=0,
tol=1e-3,
inference_cache=0,
inactive_threshold=1e-5,
inactive_window=50,
logger=None,
cache_tol='auto',
switch_to=None):
BaseSSVM.__init__(self,
model,
max_iter,
C,
verbose=verbose,
n_jobs=n_jobs,
show_loss_every=show_loss_every,
logger=logger)
self.negativity_constraint = negativity_constraint
self.check_constraints = check_constraints
self.break_on_bad = break_on_bad
self.tol = tol
self.cache_tol = cache_tol
self.inference_cache = inference_cache
self.inactive_threshold = inactive_threshold
self.inactive_window = inactive_window
self.switch_to = switch_to
self.qp_time = 0
self.inference_time = 0
self.inference_calls = 0
self.iterations_done = 0
def __init__(self,
model,
max_iter=100,
C=1.0,
verbose=0,
momentum=0.0,
learning_rate='auto',
n_jobs=1,
show_loss_every=0,
decay_exponent=1,
break_on_no_constraints=True,
logger=None,
batch_size=None,
decay_t0=10,
averaging=None,
shuffle=False,
check_every=1):
BaseSSVM.__init__(self,
model,
max_iter,
C,
verbose=verbose,
n_jobs=n_jobs,
show_loss_every=show_loss_every,
logger=logger)
self.averaging = averaging
self.break_on_no_constraints = break_on_no_constraints
self.momentum = momentum
self.learning_rate = learning_rate
self.t = 0
self.decay_exponent = decay_exponent
self.decay_t0 = decay_t0
self.batch_size = batch_size
self.shuffle = shuffle
self.alpha = 0.1
self.check_every = check_every
def __init__(self,
model,
max_iter=100,
C=1.0,
check_constraints=True,
verbose=0,
negativity_constraint=None,
n_jobs=1,
break_on_bad=False,
show_loss_every=0,
batch_size=10,
tol=1e-3,
inactive_threshold=1e-5,
inactive_window=50,
logger=None,
switch_to=None):
BaseSSVM.__init__(self,
model,
max_iter,
C,
verbose=verbose,
n_jobs=n_jobs,
show_loss_every=show_loss_every,
logger=logger)
self.negativity_constraint = negativity_constraint
self.check_constraints = check_constraints
self.break_on_bad = break_on_bad
self.batch_size = batch_size
self.tol = tol
self.inactive_threshold = inactive_threshold
self.inactive_window = inactive_window
self.switch_to = switch_to
self.w = np.ones(6)
