Python Pandas实现数据分组求平均值并填充nan

DataFrame数据格式

fillna方式实现

factordatafillna = pd.DataFrame()
industrys = newfactordata1.industryName1.unique()
for ind in industrys:
t = newfactordata1.industryName1 == ind
a = newfactordata1[t].fillna(newfactordata1[t].mean())
factordatafillna = factordatafillna.append(a)

groupby方式实现

df = pd.DataFrame({'code':[1,2,3,4,5,6,7,8],
'value':[np.nan,5,7,8,9,10,11,12],
'value2':[5,np.nan,7,np.nan,9,10,11,12],
'indstry':['农业1','农业1','农业1','农业2','农业2','农业4','农业2','农业3']},
columns=['code','value','value2','indstry'],
index=list('ABCDEFGH'))

# 只留下需要处理的列
cols = [col for col in df.columns if col not in['code','indstry']]
# 分组的列
gp_col = 'indstry'
# 查询nan的列
df_na = df[cols].isna()
# 根据分组计算平均值
df_mean = df.groupby(gp_col)[cols].mean()

print(df)

# 依次处理每一列
for col in cols:
na_series = df_na[col]
names = list(df.loc[na_series,gp_col])

t = df_mean.loc[names,col]
t.index = df.loc[na_series,col].index

# 相同的index进行赋值
df.loc[na_series,col] = t

print(df)

code  value  value2 indstry
A     1    NaN     5.0     农业1
B     2    5.0     NaN     农业1
C     3    7.0     7.0     农业1
D     4    8.0     NaN     农业2
E     5    9.0     9.0     农业2
F     6   10.0    10.0     农业4
G     7   11.0    11.0     农业2
H     8   12.0    12.0     农业3
code  value  value2 indstry
A     1    6.0     5.0     农业1
B     2    5.0     6.0     农业1
C     3    7.0     7.0     农业1
D     4    8.0    10.0     农业2
E     5    9.0     9.0     农业2
F     6   10.0    10.0     农业4
G     7   11.0    11.0     农业2
H     8   12.0    12.0     农业3