尝试：

def fn(x):
    mx, mn = x["numOfItemsPurchased"].idxmax(), x["numOfItemsPurchased"].idxmin()
    return pd.Series(
        {
            "maxPurchaseDate": x.at[mx, "purchaseDate"],
            "maxPurchaseAmount": x.at[mx, "numOfItemsPurchased"],
            "leastPurchaseDate": x.at[mn, "purchaseDate"],
            "leastPurchaseAmount": x.at[mn, "numOfItemsPurchased"],
        }
    )

# convert to datetime if necessary:
df["purchaseDate"] = pd.to_datetime(df["purchaseDate"], dayfirst=True)

out = df.groupby("ID", sort=False).apply(fn).reset_index()
print(out)

图纸：

ID maxPurchaseDate  maxPurchaseAmount leastPurchaseDate  leastPurchaseAmount
0  12      2023-01-12                 34        2023-10-12                    2
1  56      2020-03-24                 12        2020-03-24                   12
2  23      2012-06-23                  4        2012-12-23                    1
3  24      2023-10-12                 24        2023-10-12                   24
4  38      2012-02-23                 21        2012-02-23                   21
5  16      2023-10-12                 34        2023-10-12                   34
6  54      2020-03-24                 19        2020-09-02                   11