def predict(self, data):
o = self.source_learner.predict(data)
I_target = data.is_target
if self.opt_succeeded:
assert not array_functions.has_invalid(self.g)
assert not array_functions.has_invalid(self.h)
b = data.R_ul.dot(self.h)
w = data.R_ul.dot(self.g)
y_old = o.fu[I_target]
y_new = (y_old - b) / w
I_invalid = array_functions.is_invalid(y_new)
y_new[I_invalid] = y_old[I_invalid]
o.fu[I_target] = y_new
o.y[I_target] = y_new
o.b = b
o.w = w
else:
o.b = np.zeros(I_target.sum())
o.w = np.ones(I_target.sum())
o.x = data.x[I_target,:]
o.assert_input()
return o
def assert_input(self):
assert not array_functions.has_invalid(self.fu)
assert not array_functions.has_invalid(self.y)
assert not array_functions.has_invalid(self.true_y)
def train(self, data):
assert data.is_regression
y_s, y_true = self.get_predictions(data)
I_target = data.is_target
I_target_labeled = data.is_target & data.is_labeled & data.is_train
y_s = data.y[I_target_labeled]
y_true = data.true_y[I_target_labeled]
x = array_functions.standardize(data.x)
x_t = x[I_target]
x_tl = x[I_target_labeled]
C = self.C
C2 = self.C2
W_ll = array_functions.make_rbf(x_tl, self.sigma, self.metric)
W_ll_reg_inv = np.linalg.inv(W_ll+C2*np.eye(W_ll.shape[0]))
W_ul = array_functions.make_rbf(x_t, self.sigma, self.metric, x2=x_tl)
R_ll = W_ll.dot(W_ll_reg_inv)
R_ul = W_ul.dot(W_ll_reg_inv)
assert not array_functions.has_invalid(R_ll)
assert not array_functions.has_invalid(R_ul)
reg = lambda gh: SMSTransfer.reg(gh, R_ul)
#f = lambda gh: SMSTransfer.eval(gh, R_ll, R_ul, y_s, y_true, C, reg)
f = SMSTransfer.eval
jac = SMSTransfer.gradient
g0 = np.zeros((R_ll.shape[0] * 2, 1))
gh_ids = np.zeros(g0.shape)
gh_ids[R_ll.shape[0]:] = 1
maxfun = np.inf
maxitr = np.inf
constraints = []
options = {
'disp': False,
'maxiter': maxitr,
'maxfun': maxfun
}
method = 'L-BFGS-B'
#R_ll = np.eye(R_ll.shape[0])
#R_ul = np.eye(R_ll.shape[0])
#y_s = 1*np.ones(y_s.shape)
#y_true = 1*np.ones(y_s.shape)
args = (R_ll, R_ul, y_s, y_true, C, reg)
results = optimize.minimize(
f,
g0,
method=method,
jac=jac,
options=options,
constraints=constraints,
args=args
)
check_results = False
if check_results:
results2 = optimize.minimize(
f,
g0,
method=method,
jac=None,
options=options,
constraints=constraints,
args=args
)
print self.params
scipy_opt_methods.compare_results(results, results2, gh_ids)
diff = results.x-results2.x
print results.x
print results2.x
g, h = SMSTransfer.unpack_gh(results.x, R_ll.shape[0])
self.opt_succeeded = results.success
if not results.success:
print 'SMS Opt failed'
data.R_ul = R_ul
self.g = g
self.h = h
#assert results.success
pass
def assert_input(self):
assert not array_functions.has_invalid(self.fu)
assert not array_functions.has_invalid(self.y)
assert not array_functions.has_invalid(self.true_y)
