pandas 是否可以使用“NamedAgg”方法進行過濾？這是我的示例代碼：df = pd.DataFrame({'Person': ['John','Paul','John','Paul','Taylor'], 'animal': ['cat', 'dog', 'cat', 'dog','dog'], 'from' : ['breeder','adoption','adoption','breeder','wild'], 'height': [9.1, 6.0, 9.5, 34.0,55], 'weight': [7.9, 7.5, 9.9, 198.0,200]})df.groupby(['Person']).agg( number_of_animal = pd.NamedAgg(column = 'animal', aggfunc = 'count'), number_of_from = pd.NamedAgg(column = 'from', aggfunc = 'count'), total_height = pd.NamedAgg(column = 'height', aggfunc = 'sum'), total_weight = pd.NamedAgg(column = 'weight', aggfunc = 'sum') )result = pd.DataFrame({'Person': ['John','Paul','Taylor'], 'number_of_animal':[2,0,0], 'number_of_from': [1,1,0], 'total_height':[0,34,55], 'total_weight':[17.8,205.5,200]})對于每個單獨的列，我想應(yīng)用一個過濾器，例如我想過濾“number_of_animal”df['animal'] == 'cat'和“total_hight”df['height'] > 10和 number_of_fromdf['from'] == 'breeder
1 回答
一只名叫tom的貓
TA貢獻1906條經(jīng)驗 獲得超3個贊
使用DataFrame.assignfor reasign 將不匹配的值替換為NaN中的 s?Series.where：
df1 = (df.assign(animal = df['animal'].where(df['animal'] == 'cat'),
? ? ? ? ? ? ? ? ?height = df['height'].where(df['height'] > 10),
? ? ? ? ? ? ? ? ?from1 = df['from'].where(df['from'] == 'breeder')
? ? ? ? ? ? ? ? )
? ? ? ? .groupby(['Person']).agg(
? ? ? ? ? ? ? ? ?number_of_animal = pd.NamedAgg(column = 'animal', aggfunc = 'count'),
? ? ? ? ? ? ? ? ?number_of_from = pd.NamedAgg(column = 'from1', aggfunc = 'count'),
? ? ? ? ? ? ? ? ?total_height = pd.NamedAgg(column = 'height', aggfunc = 'sum'),
? ? ? ? ? ? ? ? ?total_weight = pd.NamedAgg(column = 'weight', aggfunc = 'sum')
? ? ))
print (df1)
? ? ? ? number_of_animal? number_of_from? total_height? total_weight
Person? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ??
John? ? ? ? ? ? ? ? ? ?2? ? ? ? ? ? ? ?1? ? ? ? ? ?0.0? ? ? ? ? 17.8
Paul? ? ? ? ? ? ? ? ? ?0? ? ? ? ? ? ? ?1? ? ? ? ? 34.0? ? ? ? ?205.5
Taylor? ? ? ? ? ? ? ? ?0? ? ? ? ? ? ? ?0? ? ? ? ? 55.0? ? ? ? ?200.0
添加回答
舉報
0/150
提交
取消