Data Analysis From Scratch With Python: Step By Step Guide

plt.legend()
plt.show()
# Visualising the Test set results
from matplotlib.colors import ListedColormap
X_set, y_set = X_test, y_test
X1, X2 = np.meshgrid(np.arange(start = X_set[:, 0].min() - 1, stop = X_set[:, 0].max() + 1, step =
0.01),
np.arange(start = X_set[:, 1].min() - 1, stop = X_set[:, 1].max() + 1, step = 0.01))
plt.contourf(X1, X2, classifier.predict(np.array([X1.ravel(), X2.ravel()]).T).reshape(X1.shape),
alpha = 0.75, cmap = ListedColormap(('red', 'green')))
plt.xlim(X1.min(), X1.max())
plt.ylim(X2.min(), X2.max())
for i, j in enumerate(np.unique(y_set)):
plt.scatter(X_set[y_set == j, 0], X_set[y_set == j, 1],
c = ListedColormap(('red', 'green'))(i), label = j)
plt.title('Random Forest Classification (Test set)')
plt.xlabel('Age')

---

plt.ylabel('Estimated Salary')
plt.legend()
plt.show()
When we run the code, we’ll see the following: 
Notice the similarities between the Decision Tree and Random Forest. After all,
they take a similar approach of breaking down a dataset into smaller subsets. The
difference is that Random Forest uses randomness and averaging different
decision trees to come up with a more accurate model.

---

12. Clustering
In the previous chapters, we’ve discussed Supervised Learning (Regression &
Classification). We’ve learned about learning from “labelled” data. There were
already correct answers and our job back then was to learn how to arrive at those
answers and apply the learning to new data.
But in this chapter it will be different. That’s because we’ll be starting with
Unsupervised Learning wherein there were no correct answers or labels given. In
other words, there’s only input data but there’s no output. There’s no supervision
when learning from data.
In fact, Unsupervised Learning is said to embody the essence of Artificial
Intelligence. That’s because there’s not much human supervision or intervention.
As a result, the algorithms are left on their own to discover things from data.
This is especially the case in Clustering wherein the goal is to reveal organic
aggregates or “clusters” in data.