從我的 df 開始，我想按如下方式刪除行：每次ID檢查是否至少一次COD1=COD2且 ifGRADE包含字符串'A'or 'C'。如果滿足這兩個(gè)條件，則刪除行。我的想法是這樣提出我的條件：indexNames = df[ (df[df.COD1 == df.COD2]) & (df[df['Grade'].isin(['A','C'])]) ].index然后用df.drop(indexNames , inplace=True)然而對(duì)于第一行我得到了TypeError: unsupported operand type(s) for &: 'int' and 'float'。我真的很感激任何關(guān)于如何實(shí)現(xiàn)我的目標(biāo)的想法。我的 DF：# -- create a dataframe list_columns = ['ID', 'COD1', 'COD2','Grade','Data']list_data = [ [3215575, 'A21', 'A21','A','2020-09-16 13:25:00'], [3215575, 'A11', 'A21','D','2020-09-16 13:55:00'], [3215575, 'A21', 'A21','E','2020-09-16 13:12:00'], [4058770, 'D10', '213','E','2020-09-16 12:25:00'], [4058770, '313', '313','E','2020-09-16 10:25:00'], [4058770, '313', '125','B','2020-09-16 13:19:00'], [4058771, 'C55', 'C55','C','2020-09-16 14:25:00'], [3215577, 'A21', 'A21','B','2020-09-16 13:25:00'], [3215577, 'A01', 'A21','D','2020-09-16 13:55:00'] ]df = pd.DataFrame(columns=list_columns, data=list_data)預(yù)期輸出： list_columns = ['ID', 'COD1', 'COD2','Grade','Data'] list_data = [ [4058770, 'D10', '213','E','2020-09-16 12:25:00'], [4058770, '313', '313','E','2020-09-16 10:25:00'], [3215577, 'A21', 'A21','B','2020-09-16 13:25:00'], [3215577, 'A01', 'A21','D','2020-09-16 13:55:00'] ]df2 = pd.DataFrame(columns=list_columns, data=list_data)
1 回答
慕桂英4014372
TA貢獻(xiàn)1871條經(jīng)驗(yàn) 獲得超13個(gè)贊
IIUC，，你需要一個(gè)條件，用 groupby 并用 any 進(jìn)行變換，然后反轉(zhuǎn)：
cond = (df.COD1 == df.COD2) & df['Grade'].isin(['A','C'])
df2 = df[~cond.groupby(df['ID']).transform('any')].reset_index(drop=True)
ID COD1 COD2 Grade Data
0 4058770 D10 213 E 2020-09-16 12:25:00
1 4058770 313 313 E 2020-09-16 10:25:00
2 4058770 313 125 B 2020-09-16 13:19:00
3 3215577 A21 A21 B 2020-09-16 13:25:00
4 3215577 A01 A21 D 2020-09-16 13:55:00
添加回答
舉報(bào)
0/150
提交
取消