lundi 29 avril 2019

How to solve "ValueError: Supported target types are: ('binary', 'multiclass'). Got 'multiclass-multioutput' instead." error in pyton 3.7

I want to create a program to do emotion classification using TF-IDF and SVM. Before classify the data, I have to split a dataset into data training and testing using stratified KFold. I've used the numpy array to store the texts (X) and labels (Y)

But it ended up with this error :

'ValueError: Supported target types are: ('binary', 'multiclass'). Got 'multiclass-multioutput' instead.'


this code is running on python 3.7

this is my code :

labels = []

with open(path, encoding='utf-8') as in_file:
    data = csv.reader(in_file)
    for line in data:
        labels.append(line[1])

label_np = np.array(labels)
lp = label_np.reshape(20,20)
# lp = label_np.transpose(0)
# print(lp)

result_preprocess_np = np.array(result_preprocess)
hp = result_preprocess_np.reshape(20,20)

model = LinearSVC(multi_class='crammer_singer')
total_svm = []

total_mat_svm = np.zeros((20,20))

kf = StratifiedKFold(n_splits=3)
kf.get_n_splits(hp, lp)

for train_index, test_index in kf.split(hp, lp):
    # print('Train : ', test_index, 'Test : ', test_index)
    x_train, x_test = hp[train_index], hp[test_index]
    y_train, y_test = lp[train_index], lp[test_index]


vectorizer = TfidfVectorizer(min_df=5,
                             max_df=0.8,
                             sublinear_tf=True,
                             use_idf=True)
train_vector = vectorizer.fit_transform(x_train)
test_vector = vectorizer.transform(x_test)

model.fit(x_train, y_train)
result_svm = model.score(x_test, y_test)
print(result_svm)

I expect the result is the accuracy of classification.

Aucun commentaire:

Enregistrer un commentaire