Question

当我尝试对齐数据框时，我得到了错误的结果。但我没有得到错误：

dateparse = lambda x: pd.datetime.strptime(x, '%d.%m.%Y')
#well_dates_up_obj=fhist_grouped_hist_data.reset_index(drop=True).set_index(['well_alias','Дата','upper_obj'], drop=False)
well_dates_up_obj=fhist_grouped_hist_data.copy(deep=True)

presdf = reading_csv_files(files_dict, 'davl','cp1251', sep=';' , skipinitialspace=True, header = 0,parse_dates = ['DAT'], date_parser=dateparse)
presdf['well_alias'] = presdf.apply(lambda x: renamefunc(x['NSKV'].upper()), axis = 1);
presdf = presdf.loc[presdf['well_alias'].isin(wells_im_model),:]
presdf.replace({'PPL':0, 'PZAB':0},np.nan, inplace=True)
presdf['area'] = ''
for wellname in set(presdf['well_alias']):
    presdf.loc[presdf['well_alias']==wellname,'area'] = wells[wellname].area
presdf=presdf.rename({'GOR':'davlname'},axis='columns')
presdf = presdf.set_index(['area','davlname'], drop=False)
objectsdavl_indxed = objectsdavl.set_index(['area','davlname'], drop=False)
presdf = presdf.reindex(columns = np.append( presdf.columns.values, ['woc', 'liqden', 'obj_alias']))

presdf.loc [：，['woc'，'liqden'，'obj_alias']] = objectsdavl_indxed [['woc'，'den'，'alias']]。align（presdf，join ='right' ，axis = 0）[0] .to_numpy（）

我得到了具有不同索引的错误对齐的数据框。

下面的代码是数据帧的示例：

presdf[['woc', 'liqden', 'obj_alias']].head()
Out[127]: 
                                woc  liqden obj_alias
area                        davlname                         
AREA1                       C1    980.0    0.92  C2
                            C1    980.0    0.92  C2
                            C1    980.0    0.92  C2
                            C1    980.0    0.92  C2
                            C1    980.0    0.92  C2


objectsdavl_indxed[['woc', 'den', 'alias']].align(presdf, join='right', axis=0)[0].head()
Out[128]: 
                        woc   den     alias
area                 davlname                       
AREA2                C2  980.0  0.92 C2
                     C2  980.0  0.92 C2
                     C2  980.0  0.92 C2
                     C2  980.0  0.92 C2
                     C2  980.0  0.92 C2

但是我希望得到以下结果：

objectsdavl_indxed[['woc', 'den', 'alias']].align(presdf, join='right', axis=0)[0].head()
Out[128]: 
                                woc  liqden obj_alias
area                        davlname                         
AREA1                       C1    980.0    0.92  C2
                            C1    980.0    0.92  C2
                            C1    980.0    0.92  C2
                            C1    980.0    0.92  C2
                            C1    980.0    0.92  C2

数据框对齐在熊猫中不起作用

0 个答案: