TypeError in chapter 3 with python 3.7

[rrcook]

MacOS 10.14 Mojave, python 3.7.
The cell

#cat_to_num(data['Sex'])
features = prepare_data(data_train)
features[:5]

Gives

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-6-1224e570c6ab> in <module>
      1 #cat_to_num(data['Sex'])
----> 2 features = prepare_data(data_train)
      3 features[:5]

<ipython-input-5-07c5dd9cc834> in prepare_data(data)
     16 
     17     # Adding Embarked categorical value
---> 18     features = features.join( cat_to_num(data['Embarked']) )
     19 
     20     return features

<ipython-input-4-a227bdfcf37c> in cat_to_num(data)
      2 # Changed to automatically add column names
      3 def cat_to_num(data):
----> 4     categories = unique(data)
      5     features = {}
      6     for cat in categories:

/usr/local/lib/python3.7/site-packages/numpy/lib/arraysetops.py in unique(ar, return_index, return_inverse, return_counts, axis)
    262     ar = np.asanyarray(ar)
    263     if axis is None:
--> 264         ret = _unique1d(ar, return_index, return_inverse, return_counts)
    265         return _unpack_tuple(ret)
    266 

/usr/local/lib/python3.7/site-packages/numpy/lib/arraysetops.py in _unique1d(ar, return_index, return_inverse, return_counts)
    310         aux = ar[perm]
    311     else:
--> 312         ar.sort()
    313         aux = ar
    314     mask = np.empty(aux.shape, dtype=np.bool_)

TypeError: '<' not supported between instances of 'float' and 'str'

[rrcook]

I've found that the problem is there are two nulls in the Embarked column of data/titanic.csv. You can either fill them in or add a fillna in the code, such as

    # Adding Embarked categorical value
    features = features.join( cat_to_num(data['Embarked'].fillna('Q')) )