按其他列的值创建新列组

Question

我有以下数据

df1 = pd.DataFrame({'sentence': ['A', "A", "A", "A", 'A', 'B', "B", 'B'], 'entity': ['Stay home', "Stay home", "WAY", "WAY", "Stay home", 'Go outside', "Go outside", "purpose"], 'token' : ['Severe weather', "raining", "smt", "SMT0", "Windy", 'Sunny', "Good weather", "smt"]
})


    sentence        entity      token
0   A               Stay home   Severe weather
1   A               Stay home   raining
2   A               Way         smt
3   A               Way         SMT0
4   A               Stay home   Windy
5   B               Go outside  Sunny
6   B               Go outside  Good weather
7   B               Purpose     smt

我想group by sentences的值，并在Way和Purpose存在于entity列时创建新的columns。

预期成果：

   sentence entity      token                          Way       Purpose
0   A        Stay home  Severe weather, raining, Windy smt, SMTO Nan
1   B        Go outside Sunny, Good weather            Nan       smt

Answer 1

用Series.isin在boolean indexing中为非匹配行筛选行，~用于反向掩码，聚合join，并使用DataFrame.join进行与DataFrame.pivot_table匹配的筛选行列表

vals = ['WAY','purpose']

m = df1['entity'].isin(vals)

df2 = df1[m].pivot_table(index='sentence',columns='entity',values='token', aggfunc=','.join)
df3 = df1[~m].groupby(['sentence','entity'])['token'].agg(', '.join).reset_index()

df = df3.join(df2, on='sentence')
print (df)
  sentence      entity                           token       WAY purpose
0        A   Stay home  Severe weather, raining, Windy  smt,SMT0     NaN
1        B  Go outside             Sunny, Good weather       NaN     smt