Deciding the Number of Clusters

Well done! Let's look one more time at all the dendrograms for the weather data.

As you can see, the single linkage method's dendrogram is unreadable. The average linkage method most likely led us to three clusters (if you draw a horizontal line between 75 and 100 you will intersect one blue line, and two green. The complete and ward linkages methods lead us to 4 clusters. For complete linkage, you can draw the horizontal line between 120 and 150 (it will intersect two orange and two green lines), and between 400 and 600 for ward linkage. Let's see what will be the results of using three clusters with average linkage.

Note, that in the previous sections we considered the cases of 5 or 4 clusters. Let's see how it will work now.

Aufgabe

Swipe to start coding

Import the AgglomerativeClustering function from sklearn.cluster.
Create AgglomerativeClustering model object named model with 3 clusters and using 'average' linkage.
Fit the numerical data (columns 3 - 14) to model and predict the labels. Save predicted labels as the 'prediction' column of data.
For modified DataFrame monthly_data group the observations of columns from col by 'prediction' column, and calculate the mean within each group.
Build line plot 'Month' vs 'Temp' for each value of 'Group' using monthly_data DataFrame.

Lösung

# Import the librarires

import pandas as pd

import seaborn as sns

import matplotlib.pyplot as plt

from sklearn.cluster import AgglomerativeClustering

# Read the data

data = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/138ab9ad-aa37-4310-873f-0f62abafb038/Cities+weather.csv', index_col = 0)

# Create the model

model = AgglomerativeClustering(n_clusters = 3, linkage = 'average')

# Fit the data and predict the labels

data['prediction'] = model.fit_predict(data.iloc[:,2:14])

# Extract the list of the columns

col = list(data.columns[2:14])

col.append('prediction')

# Calculate the monthly mean averages for each cluster

monthly_data = data[col].groupby('prediction').mean().stack().reset_index()

# Assign new column names

monthly_data.columns = ['Group', 'Month', 'Temp']

# Visualize the results

sns.lineplot(x = 'Month', y = 'Temp', hue = 'Group', data = monthly_data)

plt.show()

War alles klar?

Danke für Ihr Feedback!

Abschnitt 3. Kapitel 6