Find annual average of pandas dataframe with date column

user_308827 0
Viewed 11709
        id      vi       dates     f_id
0  5532714  0.549501  2015-07-07    ff_22
1  5532715  0.540969  2015-07-08    ff_22
2  5532716  0.531477  2015-07-09    ff_22
3  5532717  0.521029  2015-07-10    ff_22
4  5532718  0.509694  2015-07-11    ff_22

In the dataframe above, I want to find average yearly value for each year. This does not work:

df.groupby(df.dates.year)['vi'].transform(mean)

I get this error: *** AttributeError: 'Series' object has no attribute 'year' How to fix this?

pandas python
3 Answers

Updating and completing @piRsquared's example below for recent versions of pandas (e.g. v1.1.0), using the Grouper function instead of TimeGrouper which was deprecated:

import pandas as pd
import numpy as np
tidx = pd.date_range('2010-01-01', '2013-12-31', name='dates')
np.random.seed([3,1415])
df = pd.DataFrame(dict(vi=np.random.rand(tidx.size)), tidx)
df.groupby(pd.Grouper(freq='1Y')).mean()
user_11153770
Related