Data Analysis From Scratch With Python: Step By Step Guide

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('Decision Tree Classification (Test set)')
plt.xlabel('Age')
plt.ylabel('Estimated Salary')
plt.legend()
plt.show()
The most important difference is in this block of code:
from
sklearn.tree import DecisionTreeClassifier
classifier = DecisionTreeClassifier(criterion = 'entropy', random_state = 0)
classifier.fit(X_train, y_train)
When we run the whole code (including the data
visualization), we’ll see this: 

---

Notice the huge difference compared to Logistic Regression and K-Nearest
Neighbors (K-NN). In these latter two, there are just two boundaries. But here in
our Decision Tree Classification, there are points outside the main red region
that fall inside “mini red regions.” As a result, our model was able to capture
data points that might be impossible otherwise (e.g. when using Logistic
Regression).
Random Forest Classification
Recall from the previous chapter about Regression that a Random Forest is a
collection or ensemble of many decision trees. This also applies to Classification
wherein many decision trees are used and the results are averaged.
# Random Forest Classification
# Importing the libraries
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
%matplotlib inline
# Importing the dataset
dataset = pd.read_csv('Social_Network_Ads.csv')
X = dataset.iloc[:, [2, 3]].values
y = dataset.iloc[:, 4].values