标签编码和一个热编码器中的循环

Question

我的数据集包含分类变量，所以我使用标签编码和一个热编码器，我的代码如下

can I use a loop to ensure that my code consists of lesser lines of code?

from sklearn.preprocessing import LabelEncoder, OneHotEncoder
labelencoder_X_0 = LabelEncoder()
X[:, 0] = labelencoder_X_0.fit_transform(X[:, 0])
labelencoder_X_1 = LabelEncoder()
X[:, 1] = labelencoder_X_1.fit_transform(X[:, 1])
labelencoder_X_2 = LabelEncoder()
X[:, 2] = labelencoder_X_2.fit_transform(X[:, 2])
labelencoder_X_3 = LabelEncoder()
X[:, 3] = labelencoder_X_3.fit_transform(X[:, 3])
labelencoder_X_4 = LabelEncoder()
X[:, 4] = labelencoder_X_4.fit_transform(X[:, 4])
labelencoder_X_5 = LabelEncoder()
X[:, 5] = labelencoder_X_5.fit_transform(X[:, 5])
labelencoder_X_6 = LabelEncoder()
X[:, 6] = labelencoder_X_6.fit_transform(X[:, 6])
labelencoder_X_7 = LabelEncoder()
X[:, 7] = labelencoder_X_7.fit_transform(X[:, 7])
labelencoder_X_8 = LabelEncoder()
X[:, 8] = labelencoder_X_8.fit_transform(X[:, 8])
labelencoder_X_13 = LabelEncoder()
X[:, 13] = labelencoder_X_13.fit_transform(X[:, 13])
labelencoder_X_14 = LabelEncoder()
X[:, 14] = labelencoder_X_14.fit_transform(X[:, 14])
labelencoder_X_15 = LabelEncoder()
X[:, 15] = labelencoder_X_15.fit_transform(X[:, 15])

labelencoder_y_16 = LabelEncoder()
y[:, ] = labelencoder_y_16.fit_transform(y[:, ])

onehotencoder = OneHotEncoder(categorical_features = [1])
X = onehotencoder.fit_transform(X).toarray()
X = X[:, 1:]

onehotencoder = OneHotEncoder(categorical_features = [14])
X = onehotencoder.fit_transform(X).toarray()
X = X[:, 1:]

onehotencoder = OneHotEncoder(categorical_features = [27])
X = onehotencoder.fit_transform(X).toarray()
X = X[:, 1:]

onehotencoder = OneHotEncoder(categorical_features = [29])
X = onehotencoder.fit_transform(X).toarray()
X = X[:, 1:]

onehotencoder = OneHotEncoder(categorical_features = [38])
X = onehotencoder.fit_transform(X).toarray()
X = X[:, 1:]

onehotencoder = OneHotEncoder(categorical_features = [40])
X = onehotencoder.fit_transform(X).toarray()
X = X[:, 1:]

如何使用 for 循环 来优化代码行数？？请帮忙！

Answer 1

当然可以！我建议使用字典来存储编码器

label_encoders = {}
categorical_columns = [0, 1, 2, 3]  # I would recommend using columns names here if you're using pandas. If you're using numpy then stick with range(n) instead

for column in categorical_columns:
    label_encoders[column] = LabelEncoder()
    X[column] = label_encoders[column].fit_transform(X[column])  # if numpy instead of pandas use X[:, column] instead

Answer 2

le = LabelEncoder()
le_count = 0
for col in X.columns[1:]:
    if X[col].dtype == 'object':
        if len(list(X[col].unique())) <= 2:
            le.fit(X[col])
            X[col] = le.transform(X[col])
            le_count += 1
print('{} columns were label encoded.'.format(le_count))

这应该可以对具有 2 个或更多唯一值的任何内容进行标签编码。我遇到的唯一问题是尝试将 one-hot 编码器添加到此答案中。

标签编码和一个热编码器中的循环

For loop in Label encoding and one hot encoder

python

machine-learning

one-hot-encoding

label-encoding