改变行的值

Question

我有一个数据帧df，粘贴我关心的行

                                        plan_benefits                                              value
0                    What is the overall\ndeductible?                  $2,000 Individual / $4,000 Family

我想把这个改成下面这样-

 plan_benefits                                              value
0                    medical_deductible_single                  $2,000
1                    medical_deductible_family                  $4,000

更新-： 从value列中删除了个人和家庭

我在做下面的事-

所以基本上我获取那一行，获取值，创建一个dict并将其追加回我的原始数据帧。完成后，我可以删除前一行。 这样可以吗

 medical_deductible_single = 0
    medical_deductible_family = 0
    for index, row in df_rep.iterrows():
        if row['plan_benefits'] == 'What is the overall\ndeductible?':
            list = row['value'].split('/')
            medical_deductible_single = list[0][:7]
            medical_deductible_family = list[1][:8]

    deductible_dict = {'plan_benefits': ['medical_deductible_single', 'medical_deductible_family'], 'value': [medical_deductible_single, medical_deductible_family]}
    deductible_df = pd.DataFrame(data= deductible_dict)
    df_rep = df_rep.append(deductible_df)

Answer 1

一种方法是

data = pd.DataFrame(df.value.str.split("/",expand=True).unstack().reset_index(drop=True)).rename({0:'value'},axis=1)

data['plan_benefits'] = 'medical_deductible_' + data['value'].str.split().str[1].map({'Individual':'single', 'Family':'family'})

##print(data[['plan_benefits','value']])

---

               plan_benefits               value
0  medical_deductible_single  $2,000 Individual 
1  medical_deductible_family       $4,000 Family

Answer 2

这里有一个可行的解决方案：

import pandas as pd    

elements = df.value.values[0].split('/')[1] # separates the 'value' entry into two

col_values = []
col_values.append([i for i in elements])
col_values.append(['medical_deductible_' + i for i in ['single','family']])

columns = ['plan_benefits', 'value']
df = pd.DataFrame([{colums[i]:col_values[i] for i in range(len(columns))}])

请注意，这不是很一般，因为您的问题非常特殊（您没有提供有关数据中其他行的信息，它们是否都与您在问题中编写的行相似？）。您必须稍微调整此代码，使其适用于其他不同的行