Logistic regression on One-hot encoding

Question

I have a Dataframe (data) for which the head looks like the following:

          status      datetime    country    amount    city  
601766  received  1.453916e+09    France       4.5     Paris
669244  received  1.454109e+09    Italy        6.9     Naples

I would like to predict the status given datetime, country, amount and city

Since status, country, city are string, I one-hot-encoded them:

one_hot = pd.get_dummies(data['country'])
data = data.drop(item, axis=1) # Drop the column as it is now one_hot_encoded
data = data.join(one_hot)

I then create a simple LinearRegression model and fit my data:

y_data = data['status']
classifier = LinearRegression(n_jobs = -1)
X_train, X_test, y_train, y_test = train_test_split(data, y_data, test_size=0.2)
columns = X_train.columns.tolist()
classifier.fit(X_train[columns], y_train)

But I got the following error:

could not convert string to float: 'received'

I have the feeling I miss something here and I would like to have some inputs on how to proceed. Thank you for having read so far!

Solution 1

Consider the following approach:

first let's one-hot-encode all non-numeric columns:

In [220]: from sklearn.preprocessing import LabelEncoder

In [221]: x = df.select_dtypes(exclude=['number']) \
                .apply(LabelEncoder().fit_transform) \
                .join(df.select_dtypes(include=['number']))

In [228]: x
Out[228]:
        status  country  city      datetime  amount
601766       0        0     1  1.453916e+09     4.5
669244       0        1     0  1.454109e+09     6.9

now we can use LinearRegression classifier:

In [230]: classifier.fit(x.drop('status',1), x['status'])
Out[230]: LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)

Solution 2

To do a one-hot encoding in a scikit-learn project, you may find it cleaner to use the scikit-learn-contrib project category_encoders: https://github.com/scikit-learn-contrib/categorical-encoding, which includes many common categorical variable encoding methods including one-hot.

Comments

[m-dz]

Try y_data = data['status'] == 'received', I am pretty sure LinearRegression is expecting a numeric/boolean variable here.

[Mornor]

Thanks a lot! Would you mind explaining why my initial solution did no work out?

[MaxU - stop genocide of UA]

@Mornor, you are welcome. I guess X_train[columns] and/or y_data has some string columns, hence could not convert string to float: 'received'

[Mornor]

I would like to add that your answer is partially correct. Indeed, it only LabelEncode the strings, and not one_hot encode them. That will create false results since some string will worth "more" than others.

[Juan Acevedo]

If anyone is wondering what Mornor means, this is because label encode will be numerical values. Ex: France = 0, Italy = 1, etc. That means that some cities are worth more than others. With one-hot encoding each city has the same value: Ex: France = [1, 0], Italy = [0,1]. Also don't forget to the dummy variable trap algosome.com/articles/dummy-variable-trap-regression.html.