def run(self):
X_train, X_test, y_train, y_test = data_utils.load_train_test_data(
self.data_fname)
train_features, test_features = self.vectorizer.feature_extraction(
X_train, X_test)
gbc = GradientBoostingClassifier(n_estimators=self.n_estimators)
gbc.fit(train_features, y_train)
print(gbc.score(test_features, y_test))
def run(self):
X_train, X_test, y_train, y_test = data_utils.load_train_test_data(
self.data_fname)
train_features, test_features = self.vectorizer.feature_extraction(
X_train, X_test)
mnb = MultinomialNB()
mnb.fit(train_features, y_train)
print(mnb.score(test_features, y_test))
def run(self):
X_train, X_test, y_train, y_test = data_utils.load_train_test_data(
self.data_fname)
train_features, test_features = self.vectorizer.feature_extraction(
X_train, X_test)
lr = LogisticRegression()
lr.fit(train_features, y_train)
print(lr.score(test_features, y_test))
def run(self):
X_train, X_test, y_train, y_test = data_utils.load_train_test_data(
self.data_fname)
train_features, test_features = self.vectorizer.feature_extraction(
X_train, X_test)
svc = SVC()
svc.fit(train_features, y_train)
print(svc.score(test_features, y_test))
def run(self):
X_train, X_test, y_train, y_test = data_utils.load_train_test_data(
self.data_fname)
train_features, test_features = self.vectorizer.feature_extraction(
X_train, X_test)
perceptron = Perceptron()
perceptron.fit(train_features, y_train)
print(perceptron.score(test_features, y_test))
def run(self):
X_train, X_test, y_train, y_test = data_utils.load_train_test_data(
self.data_fname)
train_features, test_features = self.vectorizer.feature_extraction(
X_train, X_test)
rf = RandomForestClassifier(n_estimators=self.n_estimators,
criterion=self.criterion)
rf.fit(train_features, y_train)
print(rf.score(test_features, y_test))
def run(self):
X_train, X_test, y_train, y_test = data_utils.load_train_test_data(
self.data_fname)
train_features, test_features = self.vectorizer.feature_extraction(
X_train, X_test)
abc = AdaBoostClassifier(base_estimator=self.base_estimator,
n_estimators=self.n_estimators)
abc.fit(train_features, y_train)
print(abc.score(test_features, y_test))
def run(self):
X_train, X_test, y_train, y_test = data_utils.load_train_test_data(
self.data_fname,
is_raw=False,
max_seq_len=self.para['max_seq_len'])
embedding_matrix = self.vectorizer.get_embedding_matrix()
max_acc, step, stop_num = 0.0, -1, 0
# Model Training
model = DAN(embedding_matrix, self.para['hidden_size'])
optimizer = torch.optim.Adam(model.parameters(),
lr=self.para['learning_rate'],
weight_decay=self.para['l2_reg'])
for i in range(self.para['epoch_num']):
train_cost, train_acc = 0.0, 0
for j in range(int(len(X_train) / self.para['batch_size'])):
data_X = X_train[j * self.para['batch_size']:(j + 1) *
self.para['batch_size']]
data_y = y_train[j * self.para['batch_size']:(j + 1) *
self.para['batch_size']]
loss, correct_num = model.forward(data_X, data_y)
train_cost += loss.item() * self.para['batch_size']
train_acc += correct_num.item()
optimizer.zero_grad()
loss.backward()
optimizer.step()
test_cost, test_acc = 0.0, 0
for j in range(int(len(X_test) / self.para['batch_size'])):
data_X = X_test[j * self.para['batch_size']:(j + 1) *
self.para['batch_size']]
data_y = y_test[j * self.para['batch_size']:(j + 1) *
self.para['batch_size']]
loss, correct_num = model.forward(data_X, data_y)
test_cost += loss.item() * self.para['batch_size']
test_acc += correct_num.item()
print(
'Epoch %d; Training Loss: %.3f; Training Acc: %.3f. Testing Loss: %.3f; Testing Acc: %.3f'
% (i, train_cost / X_train.shape[0],
train_acc / X_train.shape[0], test_cost / X_test.shape[0],
test_acc / X_test.shape[0]))
if test_acc / X_test.shape[0] > max_acc:
max_acc = test_acc / X_test.shape[0]
step = i
else:
stop_num += 1
if stop_num == self.para['stop_num']:
break
print('Best Performance: %.3f at Epoch %d' % (max_acc, step))
def run(self):
X_train, X_test, y_train, y_test = data_utils.load_train_test_data(self.data_fname)
train_features, test_features = self.vectorizer.feature_extraction(X_train, X_test)
neigh = KNeighborsClassifier(n_neighbors=self.n_neighbors)
neigh.fit(train_features, y_train)
print(neigh.score(test_features, y_test))
def run(self):
X_train, X_test, y_train, y_test = data_utils.load_train_test_data(self.data_fname)
train_features, test_features = vectorizer.feature_extraction(X_train, X_test)
dt = DecisionTreeClassifier(criterion=self.criterion)
dt.fit(train_features, y_train)
print(dt.score(test_features, y_test))