对具有相同索引名称的 CSV 的所有项目求和？ 蟒蛇，熊猫

Question

正常的 csv 文件看起来像这样

我想总结同一个国家的所有省份。 以澳大利亚为例，我希望它看起来像这样

到目前为止，代码看起来像这样

class FileProcessCovid:

   def dataLoad(self):
       data= pd.read_csv("time_series_covid19_confirmed_global.csv", index_col=0)
       #print(data)
       return data

   def delete_columns(self, data):
       noIndx = dataframe.reset_index(drop=True)
       dropped = noIndx.drop(["Lat", "Long"], axis=1)
       # del dataframe[dataframe.columns[0]]

       print(dropped)
       return dropped

   def sum_provinces(self, dropped):
    
       pass

c = FileProcessCovid()
dataframe = c.dataLoad()
dropped = c.delete_columns(dataframe)
c.sum_provinces(dropped)

任何帮助表示赞赏。 谢谢

Answer 1

curr_country = data.iloc[0][['Country/Region'] #keeping track of the country we're working with
dd = defaultdict() #empty dict
df_out = pd.DataFrame(columns=list(data.columns)) #new output DataFrame

for row in data.iterrows():
    country = row[1]['Country/Region']
    dd['Country/Region'] = country
    if country == curr_country:
        for date in data:
            if date is not 'Country/Region' or 'Province/State':
                dd[date] += row[1][date]
    else:
        df_out.append(dd, ignore_index = True) #Populating output DF w/ country's summed values
        curr_country = country
        dd = defaultdict() #New defaultdict for new country
        dd['Country/Region'] = country
        for date in data:
            if date is not 'Country/Region' or 'Province/State':
                dd[date] = row[1][date]

df_out.append(dd, ignore_index = True) #Populating output DF w/ last country's values
return df_out