rpy2 error with datetime64[ns] pandas series

问题

How can one pass datetime64[ns] data to R from rpy2? The following simple case:

import pandas as pd
import rpy2.robjects as robjs
from rpy2.robjects import pandas2ri

pandas2ri.activate()

dti = pd.date_range('2018-01-01', periods=3, freq='H')
robjs.r.summary(dti)

fails with a ValueError: ValueError: Unknown numpy array type "datetime64[ns]". with rpy2 version 3.3.4 and pandas 1.0.5.

回答1:

Instead of passing DatetimeIndex to Ras you do:

dti = pd.date_range('2018-01-01', periods=3, freq='H')
type(dti)
pandas.core.indexes.datetimes.DatetimeIndex

try to wrap your data into DataFrame object:

dti = pd.DataFrame({"DatetimeIndex": pd.date_range('2018-01-01', periods=3, freq='H')})
type(dti)
pandas.core.frame.DataFrame

The result is:

robjs.r.summary(dti)
array(['Min.   :2018-01-01 00:00:00  ', '1st Qu.:2018-01-01 00:30:00  ',
       'Median :2018-01-01 01:00:00  ', 'Mean   :2018-01-01 01:00:00  ',
       '3rd Qu.:2018-01-01 01:30:00  ', 'Max.   :2018-01-01 02:00:00  '],
      dtype='<U29')

来源：https://stackoverflow.com/questions/62599909/rpy2-error-with-datetime64ns-pandas-series