Challenge 3: Pipelines

Pipelines play a crucial role in streamlining machine learning workflows, ensuring the coherent and efficient transition of data from one processing stage to another. Essentially, a pipeline bundles together a sequence of data processing steps and modeling into a single, unified structure. The primary advantage of using pipelines is the minimization of common workflow errors, such as data leakage when standardizing or normalizing data.

Compito

Swipe to start coding

Apply data scaling to the wine dataset, and then use the KMeans algorithm for clustering wines based on their chemical properties.

Apply data standard scaling to the features of the wine dataset.
Use the KMeans algorithm to cluster wines based on their chemical properties. You need 3 clusters.
Apply the pipeline to the data

Soluzione

import seaborn as sns

import matplotlib.pyplot as plt

from sklearn.datasets import load_wine

from sklearn.preprocessing import StandardScaler

from sklearn.cluster import KMeans

from sklearn.pipeline import Pipeline

import warnings

# Ignore warnings

warnings.filterwarnings('ignore')

# Load Wine dataset

wine = load_wine()

X = wine.data

# 1. Create a pipeline that first applies standard scaling and then KMeans clustering

pipeline = Pipeline([

('scaler', StandardScaler()),

('kmeans', KMeans(n_clusters=3, random_state=0))

])

# 2. Apply the pipeline to the data

clusters = pipeline.fit_predict(X)

# Result distribution

plt.title('Cluster assignments:')

sns.countplot(x=clusters)

plt.show()

Tutto è chiaro?

Grazie per i tuoi commenti!

Sezione 7. Capitolo 3

single