Multi-Class Classification
Multi-class classification with k-NN is as easy as binary classification. We just pick the class that prevails in the neighborhood.
The KNeighborsClassifier automatically performs a multi-class classification if y has more than two features, so you do not need to change anything. The only thing that changes is the y variable fed to the .fit() method.
Now you will perform a Multi-class classification with k-NN.
Consider the following dataset:
import pandas as pd df = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/b71ff7ac-3932-41d2-a4d8-060e24b00129/starwars_multiple.csv') print(df.head())
It is the same as in the previous chapter's example, but now the target can take three values:
0 – "Hated it" (rating is less than 3/5);
1 – "Meh" (rating between 3/5 and 4/5);
2 – "Liked it" (rating is 4/5 or higher).
Let's move to classification! Well, wait, here is the reminder of the classes you will use.
And now, let's move to classification!
Swipe to start coding
Perform a classification using the KNeighborsClassifier with n_neighbors equal to 13.
- Import the
KNeighborsClassifier. - Use the appropriate class to scale the data.
- Scale the data using
.fit_transform()for training data and.transform()for new instances. - Create the
KNeighborsClassifierobject and feedX_scaledandyto it. - Predict the classes for new instances (
X_new_scaled).
Ratkaisu
Vaihda työpöytään todellista harjoitusta vartenJatka siitä, missä olet käyttämällä jotakin alla olevista vaihtoehdoistaKiitos palautteestasi!
