Issue #414 new
It happens when the data volume is large enough like (40000, 12). The shape is changed to (40000, 4) after transferring it to R. However, using numpy2ri is fine. Sample code is shown bellow.
import numpy as np import pandas as pd from rpy2.robjects import pandas2ri as pd2r pd2r.activate() data = pd.DataFrame(np.random.randn(40000, 12), columns=list('ABCD'*3), dtype='float16') rdf = pd2r.py2ri(data) ro.r.assign('df', rdf) ro.r(r'''print(dim(df))''')