sklearn LabelEncoder：TypeError：'int'和'str'的实例之间不支持'<'

Question

我想使用KNN算法对文本进行分类。而且我有.csv扩展名的数据。

如果我使用此代码打印，数据将如下所示：

# Preprocessing

X = np.array(dataset.iloc[:, :1])
y = np.array(dataset['Class'])

print("Data variabel X : ", X)
print("Data variabel y : ", y)

输出：

[['pada awalnya aku memandang gadis itu nani namanya']['dua buah melon yang subur segar']]['Pornografi''Non-Pornografi']

我分开参加培训和考试：

# Train Test Split

from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.20)

# loading library
from sklearn.neighbors import KNeighborsClassifier
from sklearn.preprocessing import LabelEncoder

# Feature Scaling
lb = LabelEncoder()  
lb.fit(X_train)

X_train = lb.transform(X_train)  
X_test = lb.transform(X_test)

print("X_train : ", X_train)
print("X_test : ", X_test)

# instantiate learning model (k = 3)
knn = KNeighborsClassifier(n_neighbors=3)

# fitting the model
knn.fit([[X_train, y_train]], [y])

# predict the response
pred = knn.predict(X_test)

# evaluate accuracy
print (accuracy_score(y_test, pred))

我收到错误消息：

    <ipython-input-223-7d80eb4ea7d1> in <module>()
      8 
      9 X_train = lb.transform(X_train)
---> 10 X_test = lb.transform(X_test)
     11 
     12 print("X_train : ", X_train)
TypeError: '<' not supported between instances of 'int' and 'str'

我的代码有什么问题？

Answer 1

尝试一下：

lb.transform(X_test.astype(str))

基本上，您需要将数据转换为一种格式。