def pred(args):
"""
dataset: tfidf
model: MultinomialNB(alpha=0.05)
cross validate f1-score: 0.1613
test f1-score: 0.16181
"""
X_train, y_train = load_train_tfidf(), load_target()
X_test = load_test_tfidf()
clf = MultinomialNB(alpha=args.alpha)
train_and_test(clf, X_train, y_train, X_test, args.output)
def pred(args):
"""
dataset: tfidfdim
model: LinearSVC(C=1.0), ovo
cross validate f1-score: 0.1723
test f1-score: 0.16922
"""
X_train, y_train = load_train_tfidfdim(), load_target()
X_test = load_test_tfidfdim()
svm = LinearSVC(C=args.C)
clf = OneVsOneClassifier(svm)
train_and_test(clf, X_train, y_train, X_test, args.output)
def cv(args):
X_train, y_train = load_train_tfidfdim(), load_target()
svm = LinearSVC(C=args.C, verbose=args.verbose)
clf = OneVsOneClassifier(svm)
cross_validate_model(clf, X_train, y_train, cv=args.fold)
def cv(args):
X_train, y_train = load_train_tfidf(), load_target()
clf = MultinomialNB(alpha=args.alpha)
cross_validate_model(clf, X_train, y_train, cv=args.fold)
def val(args):
X_train, y_train = load_train_tfidf(), load_target()
clf = MultinomialNB(alpha=args.alpha)
validate_model(clf, X_train, y_train)