是否有一种更简洁的方法可以根据字符串中的特定子字符串获取多个变量的平均值

Question

I have variables associated with a name that i want to take the mean of, based on its MainName. Noting that i have more than two MainNames as opposed to the example below, and would look messy doing all of it. So i was wondering if anyone could make this more concise? Thanks in advance!

fullname = ['MainName1,subname1','MainName1,subname2','MainName2,subname1','MainName2,subname2']
var1 = [1,5,9,4]
var2 = [2,6,1,5]
var3 = [3,7,2,6]
var4 = [4,8,3,7]

    vars = pd.DataFrame(np.column_stack([fullname,var1,var2,var3,var4]))
    vars = vars.set_index('fullname')

    meanvars = [(allvars[allvars.index.str.contains('MainName1')]).mean(),
                (allvars[allvars.index.str.contains('MainName2')]).mean()]
    MainName = ['MainName1','MainName2']

    Final = pd.DataFrame(np.column_stack([MainName,meanvars]))

Answer 1

You can use str.extract for get substrings with joined substrings from list joined by | for regex OR passed to groupby with aggregating mean :

fullname = ['MainName1,subname1','MainName1,subname2',
            'MainName2,subname1','MainName2,subname2']
var1 = [1,5,9,4]
var2 = [2,6,1,5]
var3 = [3,7,2,6]
var4 = [4,8,3,7]

df = pd.DataFrame(np.column_stack([var1,var2,var3,var4]), index=fullname)
print (df)
                    0  1  2  3
MainName1,subname1  1  2  3  4
MainName1,subname2  5  6  7  8
MainName2,subname1  9  1  2  3
MainName2,subname2  4  5  6  7

L = ['MainName1','MainName2']
idx = df.index.str.extract('('+ '|'.join(L) + ')', expand=False)
print (idx)
Index(['MainName1', 'MainName1', 'MainName2', 'MainName2'], dtype='object')

df = df.groupby(idx).mean()
print (df)
             0    1    2    3
MainName1  3.0  4.0  5.0  6.0
MainName2  6.5  3.0  4.0  5.0