Python Data Structures and Algorithms

Prediction

With the two clusters that we have obtained, we can predict the group that a new set of data might belong to.

Let's predict which group the points [[-1.4, -1.4]] and [[2.5, 2.5]] will belong to:

    sample = np.array([[-1.4, -1.4]])     print(kmean.predict(sample))     another_sample = np.array([[2.5, 2.5]])     print(kmean.predict(another_sample)) 

The output is seen as follows:

[1][0] 

At the barest minimum, we can expect the two test datasets to belong to different clusters. Our expectation is proved right when the print statement prints 1 and 0, thus confirming that our test data does indeed fall under two different clusters.