根據 pandas 中的其他列值合並具有相同列值的行

Question

我有一個 dataframe 看起來像這樣：

data = {'Id':["A", "A", "A", "B", "B","B"], 
    'extraction': ["apple", "cherry", "alfatoxin", "ethyl", "glyphosate", "pasta"],
     "entities": ["food", "food", "hazard", "hazard", "hazard", "food"]}

df = pd.DataFrame(data)

    Id  extraction   entities
 0  A   apple        food
 1  A   cherry       food
 2  A   alfatoxin    hazard
 3  B   ethyl        hazard
 4  B   glyphosate   hazard
 5  B   pasta        food

我想要的 output 是：

    Id  food               hazards
 0  A   [apple, cherry]    alfatoxin
 1  B   pasta              [ethyl, glyphosate]

我如何在 pandas 中執行此操作？ groupby 是一個不錯的選擇嗎？

Answer 1

使用帶有聚合的pivot_table表作為list ：

df.pivot_table(index='Id', columns='entities', values='extraction', aggfunc=list)

Output：

entities             food               hazard
Id                                            
A         [apple, cherry]          [alfatoxin]
B                 [pasta]  [ethyl, glyphosate]

或者，讓單個項目不作為列表：

df.pivot_table(index='Id', columns='entities', values='extraction',
               aggfunc=lambda x: list(x) if len(x)>1 else x.squeeze())

Output：

entities             food               hazard
Id                                            
A         [apple, cherry]            alfatoxin
B                   pasta  [ethyl, glyphosate]