您可以使用多个groupby来构建所需的数据框架。请参阅代码中的注解以获取解释：

data = pd.DataFrame({"key": [f"CNI-{i+1}" for i in range(5)],
                   "ticket_id": ["T1", "T2, T3", "T1, T4", "T4, T5, T6, T7", "T1, T8, T9, T10"],
                   "ticket_category": ["Reset", "Reset, Error", "Reset, Server", "Server, Server, Server, Error", "Reset, Reset, Error, Server"]
                   })

#convert strings to lists for explode
df = data.copy().set_index("key").apply(lambda x: x.str.split(", "))

#explode to get one row per ticket_id
df = df.explode(column=list(df.columns)).reset_index()

#get count of ticket_category per original row
df["count"] = df.groupby(["key","ticket_category"])["ticket_id"].transform("count")

#groupby and construct labels
df = df.groupby(["key", "ticket_category"], as_index=False, sort=False).agg({"count": "first", "ticket_id": ", ".join})
df["combined"] = df["ticket_category"] + " - " + df["count"].astype(str) + "(" + df["ticket_id"] + ")"

#get back original df structure
df = df.groupby("key")["ticket_category", "ticket_id", "combined"].agg(", ".join).reset_index()

#get back the original column from the original data
df["ticket_category"] = df["key"].map(dict(zip(data["key"],data["ticket_category"])))

>>> df
     key                ticket_category        ticket_id  \
0  CNI-1                          Reset               T1   
1  CNI-2                   Reset, Error           T2, T3   
2  CNI-3                  Reset, Server           T1, T4   
3  CNI-4  Server, Server, Server, Error   T4, T5, T6, T7   
4  CNI-5    Reset, Reset, Error, Server  T1, T8, T9, T10   

                                            combined  
0                                      Reset - 1(T1)  
1                       Reset - 1(T2), Error - 1(T3)  
2                      Reset - 1(T1), Server - 1(T4)  
3              Server - 3(T4, T5, T6), Error - 1(T7)  
4  Reset - 2(T1, T8), Error - 1(T9), Server - 1(T10)

字符串