我有一个 data.frame 看起来与此类似(除了更长并且颜色名称更多):
ff = pd.DataFrame({'OldCol':['darkbrown','lightbeige','lightbrown / beige','beige','brown','beige / cognac'], 'NewCol':['nan','nan','nan','nan','nan','nan']})
我希望 data.frame 看起来像这样:
ffnew = pd.DataFrame({'OldCol':['darkbrown','lightbeige','lightbrown / beige','beige','brown','beige / cognac'], 'NewCol':['brown','beige','beige / brown','sand','brown','sand / brown']})
我尝试了以下方法:
ff.loc[ff['OldCol'].str.contains(r'brown|cognac',na=False) & ff['NewCol'].str.contains(r'nan'), 'NewCol'] = 'brown'
ff.loc[ff['OldCol'].str.contains(r'brown|cognac',na=False) & ~ff['NewCol'].str.contains(r'nan|brown'), 'NewCol'] = ff['NewCol']+'/ brown'
ff.loc[ff['OldCol'].str.contains(r'beige|sand',na=False) & ff['NewCol'].str.contains(r'nan'), 'NewCol'] = 'beige'
ff.loc[ff['OldCol'].str.contains(r'beige|sand',na=False) & ~ff['NewCol'].str.contains(r'nan|beige'), 'NewCol'] = ff['NewCol'] +'/ beige'
在我较长的 data.frame 中,我通常会收到错误:
ValueError: cannot reindex from a duplicate axis
有人可以帮忙吗? 非常感谢!
请您参考如下方法:
index 中存在重复问题。您可以将索引的所有值替换为 reset_index到 常规索引 (0,1,2..len(df)-1)。旧值由参数 drop=True 删除:
ff.reset_index(drop=True, inplace=True)
测试:
ff = pd.DataFrame({'OldCol':['darkbrown','lightbeige','lightbrown / beige','beige','brown','beige / cognac'], 'NewCol':['nan','nan','nan','nan','nan','nan']})
ffnew = pd.DataFrame({'OldCol':['darkbrown','lightbeige','lightbrown / beige','beige','brown','beige / cognac'], 'NewCol':['brown','beige','beige / brown','sand','brown','sand / brown']})
ff.index = [0,0,2,3,4,5]
#ValueError: cannot reindex from a duplicate axis
ff.reset_index(drop=True, inplace=True)
