Iris Interactive Plots

Introduction

You will see the tutorial for the implementation of interactive plots using NNSOM

Training

Refer Iris Training to see how to train the model before plotting

Data Preparation

Just make sure you are setting the mouse_click flag to True to see interactive plots.

num1 = get_cluster_array(X[:, 0], clust)
num2 = get_cluster_array(X[:, 1], clust)
cat = count_classes_in_cluster(y, clust)

perc_sentosa = get_perc_cluster(y, 0, clust)
iris_class_counts_cluster_array = count_classes_in_cluster(y, clust)
align = np.arange(len(iris_class_counts_cluster_array[0]))
num_classes = count_classes_in_cluster(y, clust)
num_sentosa = num_classes[:, 0]

int_dict = {
    'data': X,
    'target': y,
    'clust': clust,
    'num1': num1,
    'num2': num2,
    'cat': cat,
    'topn': 5,
}

# Interactive hit histogram
fig, ax, patches, text = som.hit_hist(X, mouse_click=True, **int_dict)
plt.show()

Output:

Initial interactive window (on clicking cluster center opens a menu)

Clustered items as pie chart ( on clicking pie )

Clustered items as stem plot ( on clicking stem )

Conclusion

In this example we see the implementation of hit histogram to see whats inside the cluster. We can follow the same for all plots by just setting the flag mouse_click = True