我有以下熊貓數據框：>>>Feature name error1 error2 error3 error40 1 A overlaps overlaps overlaps overlaps1 2 B No error 2 3 C overlaps invalid invalid 3 4 D invalid overlaps overlaps我只想為每一行提供唯一的錯誤，例如：>>>Feature Name error1 error2 error3 error40 1 A overlaps 1 2 B No error 2 3 C overlaps invalid 3 4 D invalid overlaps 有什么簡單的方法可以得到這個嗎？我想也許可以計算每個值每行出現的次數，但后來我不知道如何刪除它們
2 回答
侃侃爾雅
TA貢獻1801條經驗 獲得超16個贊
想法是從列中刪除重復項error，添加DataFrame.reindex添加可能刪除的列并分配回來：
cols = df.filter(like='error').columns
df[cols] = (df[cols].apply(lambda x: pd.Series(x.unique()), axis=1)
? ? ? ? ? ? ? ? ? ? .reindex(np.arange(len(cols)), axis=1))
print (df)
? ?Feature name? ? error1? ? error2? error3? error4
0? ? ? ? 1? ? A? overlaps? ? ? ?NaN? ? ?NaN? ? ?NaN
1? ? ? ? 2? ? B? ? ? ? No? ? ?error? ? ?NaN? ? ?NaN
2? ? ? ? 3? ? C? overlaps? ?invalid? ? ?NaN? ? ?NaN
3? ? ? ? 4? ? D? ?invalid? overlaps? ? ?NaN? ? ?NaN
桃花長相依
TA貢獻1860條經驗 獲得超8個贊
嘗試用
out = pd.DataFrame(list(map(pd.unique, df.loc[:,'error1':].values)),index=df.Feature)
Out[333]:
0 1 2
Feature
1 overlaps None None
2 No error None
3 overlaps invalid None
4 invalid overlaps None
添加回答
舉報
0/150
提交
取消