def main():
print("Loading train data")
total, total_target = getdata()
print("Normalizing data")
mu = np.mean(total, axis=0)
sigma = np.mean(total, axis=0)
X_norm = (total - mu) / sigma
print("PCA")
pca = PCA(n_components=16, whiten=True)
pca.fit(X_norm)
X_pca = pca.transform(X_norm)
print("Split train")
X_train, X_test, y_train, y_test = train_test_split(
X_pca, total_target, test_size=0.25, random_state=0)
tuned_parameters = [{'kernel': ['rbf'], 'gamma': [1e-3, 1e-4],
'C': [1, 10, 100, 1000]},
{'kernel':['linear'], 'C': [1, 10, 100, 1000]}]
print("GridSearchCV")
clf = GridSearchCV(SVC(C=1), tuned_parameters, cv=3, score_func=s.bidirectional_auc, n_jobs=-1, verbose=3)
clf.fit(X_train, y_train)
print("Best parameters")
print(clf.best_estimator_)
for params, mean_score, scores in clf.cv_scores_:
print("%0.3f (+/-$0.03f) for %r"
% (mean_score, scores.std() / 2, params))
y_true, y_pred = y_test, clf.predit(X_test)
print(classification_report(y_true, y_pred))
print(s.bidirectional_auc(y_true, y_pred))
print("Saving the classifier")
data_io.save_model(clf)
