Data Analysis From Scratch With Python: Step By Step Guide

6. A Quick Example
Iris Dataset
Let’s quickly see how data analysis and machine learning work in real world
data sets. The goal here is to quickly illustrate the potential of Python and
machine learning on some interesting problems.
In this particular example, the goal is to predict the species of an Iris flower
based on the length and width of its sepals and petals. First, we have to create a
model based on a dataset with the flowers’ measurements and their
corresponding species. Based on our code, our computer will “learn from the
data” and extract patterns from it. It will then apply what it learned to a new
dataset. Let’s look at the code.
#importing the necessary libraries
from sklearn.datasets import load_iris
from sklearn import tree
from sklearn.metrics import accuracy_score
import numpy as np
#loading the iris dataset
iris = load_iris()
x = iris.data #array of the data
y = iris.target #array of labels (i.e answers) of each data entry
#getting label names i.e the three flower species
y_names = iris.target_names
#taking random indices to split the dataset into train and test
test_ids = np.random.permutation(len(x))
#splitting data and labels into train and test
#keeping last 10 entries for testing, rest for training
x_train = x[test_ids[:-10]]
x_test = x[test_ids[-10:]]
y_train = y[test_ids[:-10]]
y_test = y[test_ids[-10:]]
#classifying using decision tree
clf = tree.DecisionTreeClassifier()
#training (fitting) the classifier with the training set
clf.fit(x_train, y_train)

---

#predictions on the test dataset
pred = clf.predict(x_test)