Practicing standardization
It is dangerous to use KNN on unknown distributions blindly. Its performance suffers greatly when the feature distributions don't have the same scales. Unscaled features will skew distance calculations and thus return unrealistic anomaly scores.
A common technique to counter this is using standardization, which involves removing the mean from a feature and dividing it by the standard deviation. This has the effect of making the feature have a mean of 0 and a variance of 1.
Practice standardization on the females dataset, which has already been loaded for you.
Questo esercizio fa parte del corso
Anomaly Detection in Python
Istruzioni dell'esercizio
- Create an instance of
StandardScaler()and store it asss. - Extract feature and target arrays into
Xandy. The target is theweightkgcolumn. - Fit
StandardScaler()to X and transform it simultaneously. - Repeat the above process but preserve the column names of the
XDataFrame.
Esercizio pratico interattivo
Prova a risolvere questo esercizio completando il codice di esempio.
from sklearn.preprocessing import StandardScaler
# Initialize a StandardScaler
ss = ____
# Extract feature and target arrays
X = ____
y = ____
# Fit/transform X
X_transformed = ____
# Fit/transform X but preserve the column names
X.____ = ____