Scribd
Upload
29 views

Machine Learning: Lecture 7: Create Your First Project

This document provides instructions for creating a machine learning project to classify iris flowers using the iris dataset, which includes 150 samples described by 4 features. It outlines loading and exploring the iris data, splitting it into training and test sets, building a decision tree classifier model, and evaluating the model's accuracy on both the training and test sets. Additionally, it suggests some homework extensions including applying normalization, comparing other classifier models, and finding the best predictive model.

Uploaded by

Bisnu Sarkar
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
29 views

Machine Learning: Lecture 7: Create Your First Project

This document provides instructions for creating a machine learning project to classify iris flowers using the iris dataset, which includes 150 samples described by 4 features. It outlines loading and exploring the iris data, splitting it into training and test sets, building a decision tree classifier model, and evaluating the model's accuracy on both the training and test sets. Additionally, it suggests some homework extensions including applying normalization, comparing other classifier models, and finding the best predictive model.

Uploaded by

Bisnu Sarkar
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17

Machine Learning

Lecture 7: Create Your First Project


COURSE CODE: CSE490
2019
Course Teacher
Dr. Mrinal Kanti Baowaly
Assistant Professor
Department of Computer Science and
Engineering, Bangabandhu Sheikh
Mujibur Rahman Science and
Technology University, Bangladesh.

Email: [email protected]
Iris flower classification
Iris dataset
 150 samples
 3 labels/categories: Species of Iris (Iris setosa, Iris virginica and Iris
versicolor)
 4 features: Sepal length, Sepal width, Petal length, Petal Width in
cm
Iris dataset instances
Import libraries
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn import tree
from sklearn.metrics import accuracy_score
Load the dataset
iris_data = pd.read_csv('IRIS.csv')
Summarize the dataset
# dimensions (no. of rows & columns)
print(iris_data.shape)
# list of columns/features
print(iris_data.columns)
# peek some data
print(iris_data.head(10))
# statistical summary
print(iris_data.describe())
Specify the target variable and its
distribution
# target variable
target = iris_data['species']

# distribution of class labels or categories


print(pd.value_counts(target))
Specify the target variable and its
distribution
# target variable
target = iris_data['species']

# distribution of class labels or categories


print(pd.value_counts(target))

# alternative of finding class distribution


print(iris_data.groupby('species').size())
Split dataset into training and test data
seed = 7
train_data, test_data = train_test_split(iris_data, test_size=0.3,
random_state= 7)
# shape of the datasets
print('\nShape of training data :',train_data.shape)
print('\nShape of testing data :',test_data.shape)
# class distribution of the training data
print(pd.value_counts(train_data['species']))
# class distribution of the test data
print(pd.value_counts(test_data['species']))
Balanced split of the dataset
seed = 7
train_data, test_data = train_test_split(iris_data, test_size=0.3,
random_state=seed, stratify=target)
Separate the independent and target
variables
# separate the independent and target variables from training data
train_x = train_data.drop(columns=['species'],axis=1)
train_y = train_data['species']

# separate the independent and target variables from test data


test_x = test_data.drop(columns=['species'],axis=1)
test_y = test_data['species']
Build the model
# create a classifier object/model
model=tree.DecisionTreeClassifier()

# train the model with fit function


model.fit(train_x, train_y)
Make predictions
# make predictions on training data
predictions_train = model.predict(train_x)
print('\nTraining Accuracy :', accuracy_score(train_y,
predictions_train))

# make predictions on test data


predictions_test = model.predict(test_x)
print('\nTest Accuracy :', accuracy_score(test_y, predictions_test))
Home work for the Lab.
Apply normalization or standardization
Apply different classifiers and compare their performances
• Logistic Regression (LR)
• K-Nearest Neighbors (KNN)
• Support Vector Machines (SVM)
Find the best model for the prediction task
Some example projects
Iris classification [Link1, Link2]
Machine Learning-Let’s Get Started [Link]
Your First Machine Learning Project in Python Step-By-Step [Link]
24 Data Science Projects To Boost Your Knowledge and Skills [link]
6 Complete Machine Learning Projects [Link]

You might also like