1.1 Oracle Machine Learning for Python Known Issues

Learn about the issues you may encounter when using Oracle Machine Learning for Python and how to work around them.

The online help for oml.Datetime.replace incorrectly states that oml.integer is supported for the year, month, day, hour minute, second, and microsecond arguments

Running the help file for oml.Datetime.replace states that the oml.integer is supported for the year, month, day, hour minute, second, and microsecond arguments but it is not yet supported:

help(oml.Datetime.replace)

oml.Integer math methods fail with numeric values

Running the oml.integer math method fails with the numeric values.

import oml
DF = oml.push(pd.DataFrame(range(5), columns=['X']))
X_min = DF['X'].min()
DF['X'] - X_min

The output returns TypeError: 'other' must be a oml.Integer, int, oml.Float, or float object.

Workaround

The workaround is to cast DF['X'] to an oml.Float data type:

oml.Float(DF['X']) - X_min

The output returns [0, 1, 2, 3, 4].

1.2 Documentation Accessibility

Access to Oracle Support