Python-h2o:如何正确地指定列类型?

我正试图将pandas数据框架导入到一个h2o框架中,并指定所需的列类型。问题是我最终试图对两个数据集执行.rbind(),但有时根据某些列的值,h2o会强制它们为实数或整数,然后它们无法执行.rbind(),因为列类型不同。我想确保我可以得到两个具有相同列类型的不同数据集,这样就可以完成这些失败。

可重复的示例如下:

import pandas as pd
import h2o

my_df1 = pd.DataFrame({'a':[1,1,0,0,1],
                       'b':[1,0,.5,.2,0]})

my_df2 = pd.DataFrame({'a':[.5,.8,0,0,1],
                       'b':[1,0,.5,.2,0]})

h2o.init()
my_h2o1 = h2o.H2OFrame(my_df1)
my_h2o2 = h2o.H2OFrame(my_df2)

my_h2o1.rbind(my_h2o2)  ### This fails

### try to manually specify the column names and types
col_names = [k for k in my_h2o1.types.keys()]
col_types = [v for v in my_h2o1.types.values()]

my_h2o3 = h2o.H2OFrame(my_df2,column_names=col_names, column_types=col_types)

my_h2o1.types.values() == my_h2o3.types.values()

my_h2o1.rbind(my_h2o3)  ### This fails still