I have a dataframe like this , How can i convert it into years and every column should come with mea-CodePudding

             tavg   tmin    tmax    prcp
year                
1990-01-01  9.4     6.0     15.1    0.0
1990-01-02  17.9    7.3     24.3    0.0
1990-01-03  14.9    8.3     22.1    0.0
1990-01-04  20.9    14.6    28.2    0.0
1990-01-05  29.8    25.0    36.0    0.3

Expected output :

        tavg    tmin    tmax    prcp
year                
1990    9.2     5.0     15.1    0.0
1991    16.4    6.3     24.3    0.0
1992    12.3    8.3     22.1    0.0
1993    21.7    14.6    28.2    0.0
1994    28.6    25.0    36.0    0.3

CodePudding user response：

Firstly, you need to convert years column into datetime64[ns] type using the line, if df is your dataframe:

df["datetime"] = pd.to_datetime(df["year"], format='%Y-%m-%d')

and then, group by year as follow:

df = df.groupby(df.index.year).mean()

Actually, you can't do operations on dates without converting it into datetime64[ns] type. Good luck!

CodePudding user response：

Extract year from year column and aggregate mean:

print (df)
        year  tavg  tmin  tmax  prcp
0 1990-01-01   9.4   6.0  15.1   0.0
1 1990-01-02  17.9   7.3  24.3   0.0
2 1991-01-03  14.9   8.3  22.1   0.0
3 1991-01-04  20.9  14.6  28.2   0.0
4 1992-01-05  29.8  25.0  36.0   0.3


df = df.groupby(df['year'].dt.year).mean()
print (df)
       tavg   tmin   tmax  prcp
year                           
1990  13.65   6.65  19.70   0.0
1991  17.90  11.45  25.15   0.0
1992  29.80  25.00  36.00   0.3