def main():
prior = .5
x, y, x_t, y_t = gen_twonorm_pumil(n_p=30, n_u=200, prior_u=prior, n_t=100)
print(x)
print(len(x))
print(type(x))
# print(x[0])
# print(type(x[0]))
# print(y)
# print(type(y))
breakpoint()
param_grid = {
'prior': [prior],
'lam': np.logspace(-3, 1, 5),
'basis': ['minimax']
}
lambda_list = np.logspace(-3, 1, 5)
clf = GridSearchCV(estimator=pumil_mr.PUMIL_SL(),
param_grid=param_grid,
cv=5,
n_jobs=-1)
clf.fit(x, y)
y_h = clf.predict(x_t)
print(y_h)
print(y_t)
err = 100 * bin_clf_err(y_h, y_t, prior)
print("MR: {}%".format(err))
def test_fit(self):
prior = .5
x, y, x_t, y_t = gen_twonorm_pumil(n_p=30, n_u=200,
prior_u=prior, n_t=100)
pu_sl = pumil_mr.PUMIL_SL(prior, basis='minimax')
pu_sl.fit(x, y)
y_h = pu_sl.predict(x_t)
err = bin_clf_err(y_h, y_t, prior)
self.assertLess(err, .2)
def test_cv(self):
prior = .5
x, y, x_t, y_t = gen_twonorm_pumil(n_p=30, n_u=200,
prior_u=prior, n_t=100)
param_grid = {'prior': [prior],
'lam': np.logspace(-3, 1, 5),
'basis': ['minimax']}
lambda_list = np.logspace(-3, 1, 5)
clf = GridSearchCV(estimator=pumil_mr.PUMIL_SL(),
param_grid=param_grid,
cv=5, n_jobs=-1)
clf.fit(x, y)
y_h = clf.predict(x_t)
err = bin_clf_err(y_h, y_t, prior)
self.assertLess(err, .2)
def main():
prior = .5
x, y, x_t, y_t = gen_twonorm_pu(n_p=30, n_u=200, prior_u=prior, n_t=100)
print(x)
print(y)
lambda_list = np.logspace(-3, 1, 5)
param_grid = {'prior': [prior], 'lam': lambda_list, 'basis': ['lm']}
clf = GridSearchCV(estimator=pu_mr.PU_SL(),
param_grid=param_grid,
cv=5,
n_jobs=-1)
clf.fit(x, y)
y_h = clf.predict(x_t)
print(y_t)
print(y_h)
err = 100 * bin_clf_err(y_h, y_t, prior)
print("MR: {:.2f}%".format(err))
priorh = cpe(x_l, y_l, x_u)
clf = PNU_SL(prior=priorh, basis='lm')
params = {'eta': eta_list, 'lam': [.1]}
etah = calc_etab(np.sum(y == +1), np.sum(y == -1), priorh)
clf = GridSearchCV(estimator=clf,
param_grid=params,
scoring=make_scorer(pnu_risk, prior=priorh, eta=etah),
cv=3,
n_jobs=-1)
timer.tic("Start PNU_SL")
clf.fit(x, y)
timer.toc()
y_h = clf.predict(x_t)
print(y_h)
print(y_t)
err1 = 100 * bin_clf_err(y_h, y_t, prior=prior_u)
print("Error: {:.2f}\n".format(err1))
timer.tic("Start PNU_SL_FastCV")
clf2 = PNU_SL_FastCV(x,
y,
priorh,
eta_list,
lambda_list=[.1],
n_fold=3,
basis='lm',
nargout=1)
timer.toc()
y_h = clf2(x_t)
err2 = 100 * bin_clf_err(y_h, y_t, prior=prior_u)
print("Error: {:.2f}".format(err2))
n_trial = 20
best_err = np.inf
errs1 = np.empty(n_trial)
errs2 = np.empty((n_trial, len(eta_list)))
priors = np.empty(n_trial)
for ite in range(n_trial):
x, y, x_t, y_t = gen_twonorm_ssl(n_l, prior_l, n_u, prior_u, n_t)
x_l, y_l, x_u = x[y != 0, :], y[y != 0], x[y == 0, :]
priorh = cpe(x_l, y_l, x_u)
clf = PNU_SL(prior=priorh, lam=.1, basis='lm')
params = {'eta': eta_list}
clf = GridSearchCV(estimator=clf, param_grid=params, cv=2) #, n_jobs=-1)
clf.fit(x, y)
y_h = clf.predict(x_t)
errs1[ite] = 100*bin_clf_err(y_h, y_t, prior=prior_u)
if errs1[ite] < best_err:
best_err = errs1[ite]
best_w = clf.best_estimator_.coef_
best_x, best_y = x, y
for ite_eta, eta in enumerate(eta_list):
clf = PNU_SL(prior=priorh, eta=eta, lam=.1, basis='lm')
clf.fit(x, y)
y_h = clf.predict(x_t)
errs2[ite, ite_eta] = 100*bin_clf_err(y_h, y_t, prior=prior_u)
priors[ite] = priorh
print("Average of misclassification rates: {:.1f} ({:.2f})".format(
np.mean(errs1), np.std(errs1)/np.sqrt(n_trial)))
