pandas 熊猫连接不同的索引
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/43318515/
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 concat with different indices
提问by Hamperfait
I have three data frames that I want to concatenate, but they all have different indices. All three indices have the same length. My first df look like this:
我有三个要连接的数据框，但它们都有不同的索引。所有三个索引都具有相同的长度。我的第一个 df 看起来像这样：
Index Time_start Time_end duration value
0 5 10 5 1.0
1 10 16 6 NaN
...
39 50 53 3 NaN
The second df looks like this:
第二个 df 看起来像这样：
Index Time_start Time_end duration value
40 5 10 5 2.0
42 10 16 6 NaN
...
79 50 53 3 NaN
And the third looks exactly the same but with Index = [80..119] But time_start, Time_end and duration are exactly the same. Value differs.
第三个看起来完全一样，但是 Index = [80..119] 但是 time_start、Time_end 和 duration 完全相同。价值不同。
I want to concatenate the value column so that it looks like this
我想连接值列，使其看起来像这样
Index Time_start Time_end duration value1 value2 value3
1 5 10 5 1.0 2 3
2 10 16 6 NaN NaN NaN
...
39 50 53 3 NaN NaN NaN
So far I tried this
到目前为止我试过这个
pd.concat([df1, df2.value, ms3.value], axis=1, join_axes = [df1.index])
but indices are not the same, so it doesn't work. I know I can try first with
但索引不一样，所以它不起作用。我知道我可以先尝试
df2.reset_index(drop=True)
and then do the concat, which works, but I'm sure there's a better way.
然后进行 concat，它有效，但我相信有更好的方法。
回答by piRSquared
dfs = [df1, df2]
cols = ['Time_start', 'Time_end', 'duration']
keys = ['value1', 'value2']
pd.concat(
[df.set_index(cols).value for df in dfs],
axis=1, keys=keys)
value1 value2
Time_start Time_end duration
5 10 5 1.0 2.0
10 16 6 NaN NaN
50 53 3 NaN NaN
回答by jezrael
Use:
用：
dfs = [df1,df2]
k = ['value1','value2']
df = pd.concat([x.set_index(['Time_start','Time_end','duration']) for x in dfs],
axis=1,keys=k)
df.columns = df.columns.droplevel(-1)
print (df)
value1 value2
Time_start Time_end duration
5 10 5 1.0 2.0
10 16 6 NaN NaN
50 53 3 NaN NaN
Another solution:
另一种解决方案：
dfs = [df1,df2]
df = pd.concat([x.set_index(['Time_start','Time_end','duration']) for x in dfs],axis=1)
df.columns = [x + str(i+1) for i, x in enumerate(df.columns)]
print (df)
value1 value2
Time_start Time_end duration
5 10 5 1.0 2.0
10 16 6 NaN NaN
50 53 3 NaN NaN
