Suppose that I have this data frame:
df = pd.DataFrame({'col1': [1, 2, 3, 4, 5],
'col2': [7, 45, 12, 56, 14],
'col3': [56, 67, 8, 12, 39],
'col4': [16, np.nan, 25, 6, 19],
'col5': [1, 9, 23, 56, np.nan],
'col6': [13, 3, 53, 72, 88]})
All I want is to calculate mean of even columns and odd columns of this dataframe. I have tried these codes:
df['avg_odd'] = df[[df.columns[0],df.columns[2],df.columns[4]]].mean(axis=1)
df['avg_even'] = df[[df.columns[1],df.columns[3],df.columns[5]]].mean(axis=1)
But is there any way to do it faster? How should I calculate if I have 100 columns or more?
Create helper arange by length of columns with modulo and create new columns:
arr = np.arange(len(df.columns)) % 2
df['avg_odd'] = df.iloc[:, arr == 0].mean(axis=1)
df['avg_even'] = df.iloc[:, arr == 1].mean(axis=1)
print (df)
col1 col2 col3 col4 col5 col6 avg_odd avg_even
0 1 7 56 16.0 1.0 13 19.333333 12.000000
1 2 45 67 NaN 9.0 3 26.000000 24.000000
2 3 12 8 25.0 23.0 53 11.333333 30.000000
3 4 56 12 6.0 56.0 72 24.000000 44.666667
4 5 14 39 19.0 NaN 88 22.000000 40.333333
Using %
and groupby
df[['avg_odd', 'avg_even']] = df.groupby(np.arange(df.shape[1]) % 2, axis=1).mean()
col1 col2 col3 col4 col5 col6 avg_even avg_odd
0 1 7 56 16.0 1.0 13 12.000000 19.333333
1 2 45 67 NaN 9.0 3 24.000000 26.000000
2 3 12 8 25.0 23.0 53 30.000000 11.333333
3 4 56 12 6.0 56.0 72 44.666667 24.000000
4 5 14 39 19.0 NaN 88 40.333333 22.000000
df = df.assign(avg_even = df[df.columns[::2]].mean(axis=1),
avg_odd = df[df.columns[1::2]].mean(axis=1))
Simple and direct
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With