我需要将一些行从导入到 Python 的 CSV 更改为列。 CSV 有副标题

Question

这是我导入python的原始CSV数据

BNS	M1	M2	M3	M1	M2	M3
Creditor	GWP	GWP	GWP	Inf	Inf	Inf
Single Prem	100	200	300	600	900	50
Double Prem	200	300	400	330	100	44
Non Creditor
Rent	33	44	12	75	111	799
Spouse	90	80	86	100	10	100

有没有办法得到这样的：

Product	Months	Category	Value
Single Prem	M1	GWP	100
Single Prem	M2	GWP	200
Single Prem	M3	GWP	300
Single Prem	M1	Inf	600
Single Prem	M2	Inf	900
Single Prem	M3	Inf	50
Double Prem	M1	GWP	200
Double Prem	M2	GWP	300
Double Prem	M3	GWP	400
Double Prem	M1	INF	330
Double Prem	M2	INF	100
Double Prem	M3	INF	44

我在使用方面取得了一些进展

df.columns = df.columns.str.extract('(\d+)', expand=False)
df = df.stack().reset_index(name='M').rename(columns={'level_3':'R_no'})

但还是很远

Answer 1

我已将您的第一个表添加到名为“TestCSV.csv”的 csv 文件中。下面的代码正在生成所需的输出，您可以根据需要删除不需要的列。

from ast import If
import pandas as pd
from pandas.core.frame import DataFrame

import csv 

df = pd.read_csv('TestCSV.csv')
print(df.to_string()) 
#print(df.columns)
df = df.drop(0)
df = df.melt(id_vars = ['Product'], var_name = 'Months', value_name = 'Value')
df['Group'] = (
     df.apply(lambda x: 'GWP' if (x.Months == 'M1' or x.Months == 'M2' or x.Months == 'M3') else 'inf', axis=1)
     )
df['NewMonths'] = (
     df.apply(lambda x: x.Months[0:2] if (x.Months == 'M1.1' or x.Months == 'M2.1' or x.Months == 'M3.1') else x.Months, axis=1)
     )
print('\n')
print('Desired output....')
print('\n')
print(df)