AttributeError: module 'numpy' has no attribute 'float128' in data_container.py

[jasonlines]

When calling a.to_pandas_dataframe() where a is an XDataFrame, the following exception is thrown:

AttributeError: module 'numpy' has no attribute 'float128'

Specifically, it occurs in \xpandas\data_container\data_container.py:

~\Anaconda3\envs\sktime\lib\site-packages\xpandas\data_container\data_container.py in to_pandas_dataframe(self)
    251         is_all_columns_are_primitive = all(
    252             _is_class_a_primitive(dt)
--> 253             for dt in data_types
    254         )
    255         if is_all_columns_are_primitive:

~\Anaconda3\envs\sktime\lib\site-packages\xpandas\data_container\data_container.py in <genexpr>(.0)
    251         is_all_columns_are_primitive = all(
    252             _is_class_a_primitive(dt)
--> 253             for dt in data_types
    254         )
    255         if is_all_columns_are_primitive:

~\Anaconda3\envs\sktime\lib\site-packages\xpandas\data_container\data_container.py in _is_class_a_primitive(cls)
     30     '''
     31     primitives = [
---> 32         np.float16, np.float32, np.float64, np.float128,
     33         np.int8, np.int16, np.int32, np.int64,
     34         bool, str, np.uint8, np.uint16, np.uint32, np.uint64,

AttributeError: module 'numpy' has no attribute 'float128' 

I believe this is because float128 is platform-dependent. e.g. stack overflow suggests:

float128 exists on some platforms but not others, and on those platforms where it does exist it's almost certainly simply the 80-bit x87 extended precision type, padded to 128 bits