Pandas 在 Python 中将某些行转换为列
声明:本页面是StackOverFlow热门问题的中英对照翻译,遵循CC BY-SA 4.0协议,如果您需要使用它,必须同样遵循CC BY-SA许可,注明原文地址和作者信息,同时你必须将它归于原作者(不是我):StackOverFlow
原文地址: http://stackoverflow.com/questions/48958035/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Pandas Convert Some Rows to Columns in Python
提问by DYZ
So my dataset has some information by business n dates as below:
所以我的数据集有一些业务 n 日期的信息,如下所示:
Business Date Value
a 1/1/2017 127
a 2/1/2017 89
b 2/1/2017 122
a 1/1/2018 555
a 2/1/2018 455
I need this data as below format: How can i tranpose it . And i dont want multilevel in my output dataset
我需要以下格式的数据:我如何转置它。我不想在我的输出数据集中多级
Business 1/1/2017 2/1/2017 1/1/2018 2/1/2018
a 127 89 555 455
b N/A 122 N/A N/A
I tried below syntax:
我尝试了以下语法:
df = df.set_index(['Business','Date'])['Value'].unstack()
df=df.pivot(index='Business', columns='Date', values='Value')
i got the output as below:
我得到如下输出:
Date 1/1/2017 2/1/2017 1/1/2018 2/1/2018
Business
a 454 5555 555 444
b - 444 - -
when i print columns, it doesn't show LOB as column. My final dataframe should also include Business,Date fields as columns so that i can join this dataframe with another dataframe on business
当我打印列时,它不会将 LOB 显示为列。我的最终数据框还应包括业务、日期字段作为列,以便我可以将此数据框与业务上的另一个数据框连接起来
回答by DYZ
You are very close to what you want. All you need is to remove the custom index and replace it with the default index.
你非常接近你想要的。您所需要的只是删除自定义索引并将其替换为默认索引。
pivoted = df.pivot(index='Business', columns='Date', values='Value')\
.reset_index()
pivoted.columns.name=None
print(pivoted)
# Business 1/1/2017 1/1/2018 2/1/201 2/1/2017
#0 a 127.0 555.0 455.0 99.0
#1 b NaN NaN NaN 122.0