KNeighborsClassifier

Implementing k-Nearest Neighbors is pretty straightforward. We only need to import and use the KNeighborsClassifier class.

Once you imported the class and created a class object like this:

# Importing the class
from sklearn.neighbors import KNeighborsClassifier 
knn = KNeighborsClassifier(n_neighbors=3)

You need to feed it the training data using the .fit() method:

knn.fit(X_scaled, y)

And that's it! You can predict new values now.

y_pred = knn.predict(X_new_scaled)

Scaling the data

However, remember that the data must be scaled. StandardScaler is commonly used for this purpose:

You should calculate x̄ (mean) and s (standard deviation) on the training set using either .fit() or .fit_transform() method. This step ensures that the scaling parameters are derived from the training data.

When you have test set to predict, you must use the same x̄ and s to preprocess this data using .transform(). This consistency is crucial because it ensures that the test data is scaled in the same way as the training data, maintaining the integrity of the model's predictions.

# Importing the class
from sklearn.preprocessing import StandardScaler 

scaler = StandardScaler()
# Calculating x̄ and s and scaling `X_train`
X_train_scaled = scaler.fit_transform(X_train) 
# Scaling `X_test` with x̄ and s calculated in the previous line
X_test_scaled = scaler.transform(X_test) 

If you use different x̄ and s for training set and test set, your predictions will likely be worse.

Example

Let's explore a straightforward example where we aim to predict whether a person will enjoy Star Wars VI based on their ratings for Star Wars IV and V. The data is taken from The Movies Dataset with extra preprocessing. A person is considered to like Star Wars VI if they rate it more than 4 (out of 5).

After training our model, we'll make predictions for two individuals from the test set. The first individual rates Star Wars IV and V as 5 and 5, respectively, while the second individual rates them as 4.5 and 4.