我是 python/pandas 的新手。我正在尝试将行转置为列。对于任何误解，我们深表歉意 input_df(pic) output_df(pic)
input_df:
Date Project Processes Time_in_sec Time_measures
-----------------------------------------------------------------------
7/6/2017 FE eBanking .aspx 157 Average Response Time
9/2/2017 PCB eBanking Frontpage.fi 227 Call per hour
1/23/2017 ICC Acct Transfer.dc 28 Average Response Time
1/24/2017 PCB Transaction .com 0 Number of calls
1/23/2017 ICC eBanking Logon.no 0 Number of calls
output_df:
Date Project Processes Average Response Time Call per hour Number of calls
-----------------------------------------------------------------------------
7/6/2017 FE eBanking .aspx 157
9/2/2017 PCB eBanking Frontpage.fi 227
1/24/2017 PCB Transaction .com 0
1/23/2017 ICC Acct Transfer.dc 28
1/23/2017 ICC eBanking Logon.no 0
我尝试了下面的代码，在输出中只有 3 列“平均响应时间”、“每小时调用”、“调用次数”below_code's_output .但我很困惑如何获得所有必要的 6 列。你能帮忙吗？
output_df = input_df.pivot(columns = 'Time_measures', values= 'Time_in_sec')
最佳答案
IIUC 您可以将 pivot_table 与 reset_index 一起使用
df.pivot_table(columns='Time_measures', values='Time_in_sec',index=['Date','Project','Processes'],fill_value='').reset_index()
Out[98]:
Time_measures Date Project Processes \
0 1/23/2017 ICC Acct Transfer.dc
1 1/23/2017 ICC eBanking Logon.no
2 1/24/2017 PCB Transaction .com
3 7/6/2017 FE eBanking .aspx
4 9/2/2017 PCB eBanking Frontpage.fi
Time_measures Average Response Time Call per hour Number of calls
0 28
1 0
2 0
3 157
4 227
关于Python/Pandas - 多维数据透视，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/49180178/