一个更方便的编码分类特征的方法是使用Pandas的工具，我实现如下：

df = pd.DataFrame({'feature':['French', 'Germany', 'Germany', 'Spain']})
labelencoder = LabelEncoder()# Assigning numerical values and storing in another column
df['categorical_feature'] = labelencoder.fit_transform(df['feature'])
enc = OneHotEncoder(handle_unknown='ignore')# passing cat column (label encoded values of df)
enc_df = pd.DataFrame(enc.fit_transform(df[['categorical_feature']]).toarray(), columns=labelencoder.classes_)# merge with main df on key values
df = df.join(enc_df)
df

输出：

feature  categorical_feature  French  Germany  Spain
0   French                    0     1.0      0.0    0.0
1  Germany                    1     0.0      1.0    0.0
2  Germany                    1     0.0      1.0    0.0
3    Spain                    2     0.0      0.0    1.0