Many machine learning algorithms require all the features to be continuous variables. It means that if some of the features are category variables, we need to find a strategy to convert them into continuous variables. One-hot encoding is one of the most effective ways of performing this transformation. For this particular problem, the only category variable we have is Gender. Let's convert that into a continuous variable using one-hot encoding:

enc = sklearn.preprocessing.OneHotEncoder()
enc.fit(dataset.iloc[:,[0]])
onehotlabels = enc.transform(dataset.iloc[:,[0]]).toarray()
genders = pd.DataFrame({'Female': onehotlabels[:, 0], 'Male': onehotlabels[:, 1]})
result = pd.concat([genders,dataset.iloc[:,1:]], axis=1, sort=False)
result.head(5)

Once it's converted, let's look at the dataset again:

Notice that in order to convert a variable from a category variable into a continuous variable, one-hot encoding has converted Gender into two separate columns—Male and Female.