How Similar are the Results?

Well done! Let's look at the last line charts you built in the previous chapter.

As you can see, only the ward linkage could catch the 'downward up to July' trend. Both results are different. But let's find out how different they are using the rand index.

Task

Swipe to start coding

Compute the rand index to compare the results of using complete and ward linkages. Follow the next steps:

Import functions needed:

rand_score from sklearn.metrics.
AgglomerativeClustering from sklearn.cluster.

Create two models model_complete and model_ward performing a hierarchical clustering with 4 clusters both and 'complete' and 'ward' linkages respectively.
Fit the 3-14 columns of data to models and predict the labels. Save the labels for model_complete within labels_complete and for model_ward within labels_ward.
Compute the rand index using labels_complete and labels_ward.

Solution

# Import the libraries

import pandas as pd

import matplotlib.pyplot as plt

from sklearn.metrics import rand_score

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)

# Creating the models

model_complete = AgglomerativeClustering(n_clusters = 4, linkage = 'complete')

model_ward = AgglomerativeClustering(n_clusters = 4, linkage = 'ward')

# Fitting and predicting the labels

labels_complete = model_complete.fit_predict(data.iloc[:,2:14])

labels_ward = model_ward.fit_predict(data.iloc[:,2:14])

# Compute the Rand index

rand_index = rand_score(labels_complete, labels_ward)

print(f"The rand index for complete and ward linkages models is {rand_index}")

Everything was clear?

Thanks for your feedback!

Section 3. Chapter 8

# Import the libraries

import pandas as pd

import matplotlib.pyplot as plt

___

# 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)

# Creating the models

model_complete = ___(___, ___)

model_ward = ___(___, ___)

# Fitting and predicting the labels

labels_complete = ___.___(data.iloc[:,2:14])

labels_ward = ___.___(data.iloc[:,2:14])

# Compute the Rand index

rand_index = ___(___, ___)

print(f"The rand index for complete and ward linkages models is {rand_index}")