用于分类功能的OneHotEncoder

时间:2019-11-05 01:42:12

标签: one-hot-encoding

我已使用以下代码对我的分类功能进行OneHotEncode。但是,在运行代码并检查形状后,它只有 2列(事实并非如此)。代码有什么问题吗?

features_cat_cols = ['loan_type', 'property_type', 'loan_purpose', 'occupancy', 'preapproval', 'msa_md', 'state_code', 'county_code',
                    'applicant_ethnicity', 'applicant_race', 'applicant_sex', 'lender', 'co_applicant']

def encode_string(cat_features):
    ## Now, apply one hot encoding
    ohe = preprocessing.OneHotEncoder(categories='auto')
    encoded = ohe.fit_transform(cat_features.values.reshape(-1,1)).toarray()
    pdfn = ohe.get_feature_names()
    print(pdfn)
    return encoded

for col in features_cat_cols:
    Features = encode_string(df[col])

0 个答案:

没有答案