Hi all, is there a bug with h2o on reading any particular file/dataset? I was trying to perform modeling on a dataset, but while reading I observed that there are 2 extra rows. Curiously, I found that h2o shows 4 missing values in a particular column, where there is None. Anybody facing a similar issue?
This is a replica of the file I am testing, it has 9919 (including header) rows and 23 columns. The column beds have 0 NA values.
Below is the screenshot of the file when read using h2o and when read using pandas.
h2o introduces two more rows when compared with the original file
Interestingly, it introduces NAN values for 4 of the entries in the 'beds' column (within the same screenshot below) whereas it doesn't actually contain any null values.
Is there any particular reason for this?
Thanks for listening.