Question

有两个数据帧：

df1:
id sen                a1    a2    a3     a4
1  I am here          I     am    here  
2  You are there,Bar  You  are    there  Bar
3  take your time     take your   time

df2:                 
id  sen           b1   b2   b3
1 I am here        
2 no way for this

我需要比较f1.sen和f2.sen，如果它们相等，请使用a1，a2，a3和a4列中的值填写b1，b2和b3列。输出应该是这样的：

df2:                 
id  sen            b1   b2   b3
1   I am here      I   am    here 
2   no way for this

这是我的代码：

if df2['sen'].equals(df1['sen']):
    df2['b1'] = df1['a1']
    df2['b2'] = df1['a2']
    df2['b3'] = df1['a3']

但它不起作用。有什么建议吗？

我也用过这个：

if df2.sen == df1.loc[df2.index].sen: 
    df2['B1'] = df1['A1']
    df2['B2'] = df1['A2']
    df2['B3'] = df1['A3']

但输出是：

The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all(). How I can fix it? any suggestion?

Answer 1

你可以在这里做左联盟。

joined = df2.join(df1.set_index("sen"), how="left", on="sen")

然后将b1的值从a1分配给a3

joined.b1=joined.a1
joined.b2=joined.a2
joined.b3=joined.a3

删除a *列

for col in joined.columns:
    if col[0] == "a":
        del(joined[col])

根据其他数据框列中的值将值分配给数据框的列

1 个答案: