Description

Implement machine learning-based clustering and classification in Python for pattern recognition and data analysis

In this age of big data, companies across the globe use Python to sift through the avalanche of information at their disposal. By becoming proficient in unsupervised and supervised learning in Python, you can give your company a competitive edge and level up in your career. This course will give you a robust grounding in clustering and classification, the main aspects of machine learning. The course consists of 7 sections that will help you master Python machine learning. You'll begin with an introduction to Python data science and Anaconda, which is a powerful Python-driven framework for data science. Next, you'll delve into Pandas and read data structures, including CSV, Excel, and HTML data. As you advance, you'll perform data cleaning and munging to remove NAs\no data and discover how to handle conditional data, group by attributes, and do much more. You'll also grasp basic concepts of unsupervised learning such as K-means clustering and its implementation on the Iris dataset. The course will take you through the theory of dimension reduction and feature selection for machine learning and help you understand Principal Component Analysis (PCA) using two case studies. You'll get to grips with the linear and non-linear classification of SVM along with Gradient Boosting Machine (GBM) and Naive Bayes Classification. Finally, you'll explore neural networks and discover the powerful H20 framework and for deep learning classification. Additionally, you'll learn about perceptrons and Artificial Neural Networks (ANN) for binary classification. By the end of this course, you'll be able to use packages such as NumPy, Pandas, and Matplotlib to work with real data in Python. All code and supporting files for this course are available at https://github.com/sanjanapackt/PacktPublishing-Clustering-and-Classification-with-Machine-Learning-in-Python.

What You Will Learn

Harness the power of Anaconda/iPython for practical data science
Read data into the Python environment from different sources
Carry out basic data preprocessing and wrangling in Python
Implement unsupervised/clustering techniques such as k-means clustering
Get to grips with dimensionality reduction techniques and feature selection
Implement supervised learning/classification techniques such as random forests
Explore neural network- and deep learning-based classification

Audience

If you want to build data science applications in the Python environment, this book is for you. You'll also find this book helpful if you want to learn how to implement unsupervised learning on real data using Python, or if you're a student looking to get started with artificial neural networks and deep learning.

Approach

This course is full of interesting and illustrative examples and easy-to-understand theory to help you implement a real-world concept in every lecture. It covers hands-on methods to simplify and address even the most difficult concepts in Python using minimal jargon.

Key Features

Explore the most important Python data science concepts and packages, including Pandas * Master the Anaconda framework and use it to implement clustering and classification models on your data * Get to grips with data science fundamentals and understand which models should be used when *

Github Repo

https://github.com/packtpublishing/clustering-and-classification-with-machine-learning-in-r

About the Author

Minerva Singh

Minerva Singh is a PhD graduate from Cambridge University where she specialized in Tropical Ecology. She is also a part-time Data Scientist. As part of her research, she must carry out extensive data analysis, including spatial data analysis. For this purpose, she prefers to use a combination of freeware tools: R, QGIS, and Python. She does most of her spatial data analysis work using R and QGIS. Apart from being free, these are very powerful tools for data visualization, processing, and analysis. She also holds an MPhil degree in Geography and Environment from Oxford University. She has honed her statistical and data analysis skills through several MOOCs, including The Analytics Edge and Statistical. In addition to spatial data analysis, she is also proficient in statistical analysis, machine learning, and data mining.

Course Outline

1. INTRODUCTION TO THE COURSE: The Key Concepts and Software Tools

1. Welcome to Clustering & Classification with Machine Learning in Python

INTRODUCTION TO THE COURSE: The Key Concepts and Software Tools: Welcome to Clustering & Classification with Machine Learning in Python

2. What is Machine Learning?

INTRODUCTION TO THE COURSE: The Key Concepts and Software Tools: What is Machine Learning?

3. Python Data Science Environment

INTRODUCTION TO THE COURSE: The Key Concepts and Software Tools: Python Data Science Environment

4. For Mac Users

INTRODUCTION TO THE COURSE: The Key Concepts and Software Tools: For Mac Users

5. IPython in Browser

INTRODUCTION TO THE COURSE: The Key Concepts and Software Tools: IPython in Browser

6. Python Data Science Packages to Be Used

INTRODUCTION TO THE COURSE: The Key Concepts and Software Tools: Python Data Science Packages to Be Used

2. Read in Data from Different Sources with Pandas

1. What are Pandas?

Read in Data from Different Sources with Pandas: What are Pandas?

2. Read in Data from CSV

Read in Data from Different Sources with Pandas: Read in Data from CSV

3. Read in Online CSV

Read in Data from Different Sources with Pandas: Read in Online CSV

4. Read in Excel Data

Read in Data from Different Sources with Pandas: Read in Excel Data

5. Read in HTML Data

Read in Data from Different Sources with Pandas: Read in HTML Data

6. Read in Data from Databases

Read in Data from Different Sources with Pandas: Read in Data from Databases

3. Data Cleaning & Munging

1. Remove Missing Values

Data Cleaning & Munging: Remove Missing Values

2. Conditional Data Selection

Data Cleaning & Munging: Conditional Data Selection

3. Data Grouping

Data Cleaning & Munging: Data Grouping

4. Data Sub setting

Data Cleaning & Munging: Data sub setting

5. Ranking & Sorting

Data Cleaning & Munging: Ranking & Sorting

6. Concatenate

Data Cleaning & Munging: Concatenate

7. Merging & Joining Data Frames

Data Cleaning & Munging: Merging & Joining Data Frames

4. Unsupervised Learning in Python

1. Unsupervised Classification- Some Basic Concepts

Unsupervised Learning in Python: Unsupervised Classification- Some Basic Concepts

2. K-Means Clustering: Theory

Unsupervised Learning in Python: K-Means Clustering: Theory

3. Implement K-Means on the Iris Data

Unsupervised Learning in Python: Implement K-Means on the Iris Data

4. Quantifying K-Means Clustering Performance

Unsupervised Learning in Python: Quantifying K-Means Clustering Performance

5. K-Means Clustering with Real Data

Unsupervised Learning in Python: K-Means Clustering with Real Data

6. How to Select the Optimal Number of Clusters?

Unsupervised Learning in Python: How to Select the Optimal Number of Clusters?

7. Gaussian Mixture Modelling (GMM)

Unsupervised Learning in Python: Gaussian Mixture Modelling (GMM)

8. Hierarchical Clustering-theory

Unsupervised Learning in Python: Hierarchical Clustering-theory

9. Hierarchical Clustering-practical

Unsupervised Learning in Python: Hierarchical Clustering-practical

5. Dimension Reduction & Feature Selection for Machine Learning

1. Principal Component Analysis (PCA)-Theory

Dimension Reduction & Feature Selection for Machine Learning: Principal Component Analysis (PCA)-Theory

2. Principal Component Analysis (PCA)-Case Study 1

Dimension Reduction & Feature Selection for Machine Learning: Principal Component Analysis (PCA)-Case Study 1

3. Principal Component Analysis (PCA)-Case Study 2

Dimension Reduction & Feature Selection for Machine Learning: Principal Component Analysis (PCA)-Case Study 2

4. Linear Discriminant Analysis (LDA) for Dimension Reduction

Dimension Reduction & Feature Selection for Machine Learning: Linear Discriminant Analysis (LDA) for Dimension Reduction

5. t-SNE Dimension Reduction

Dimension Reduction & Feature Selection for Machine Learning: t-SNE Dimension Reduction

6. Feature Selection to Select the Most Relevant Predictors

Dimension Reduction & Feature Selection for Machine Learning: Feature Selection to Select the Most Relevant Predictors

7. Recursive Feature Elimination (RFE)

Dimension Reduction & Feature Selection for Machine Learning: Recursive Feature Elimination (RFE)

6. Supervised Learning: Classification

1. Concepts Behind Supervised Learning

Supervised Learning: Classification: Concepts Behind Supervised Learning

2. Data Preparation for Supervised Learning

Supervised Learning: Classification: Data Preparation for Supervised Learning

3. Pointers on Evaluating the Accuracy of Classification Modelling

Supervised Learning: Classification: Pointers on Evaluating the Accuracy of Classification Modelling

4. Using Logistic Regression as a Classification Model

Supervised Learning: Classification: Using Logistic Regression as a Classification Model

5. kNN- Classification

Supervised Learning: Classification: kNN- Classification

6. Naive Bayes Classification

Supervised Learning: Classification: Naive Bayes Classification

7. Linear Discriminant Analysis

Supervised Learning: Classification: Linear Discriminant Analysis

8. SVM- Linear Classification

Supervised Learning: Classification: SVM- Linear Classification

9. Non-Linear SVM Classification

Supervised Learning: Classification: Non-Linear SVM Classification

10. RF-Classification

Supervised Learning: Classification: RF-Classification

11. Gradient Boosting Machine (GBM)

Supervised Learning: Classification: Gradient Boosting Machine (GBM)

12. Voting Classifier

Supervised Learning: Classification: Voting Classifier

7. Neural Networks and Deep Learning Based Classification Techniques

1. Perceptrons for Binary Classification

Neural Networks and Deep Learning Based Classification Techniques: Perceptrons for Binary Classification

2. Artificial Neural Networks (ANN) for Binary Classification

Neural Networks and Deep Learning Based Classification Techniques: Artificial Neural Networks (ANN) for Binary Classification

3. Multi-class Classification With MLP

Neural Networks and Deep Learning Based Classification Techniques: Multi-class Classification With MLP

4. Introduction to H20

Neural Networks and Deep Learning Based Classification Techniques: Introduction to H20

5. Use H20 for Deep Learning Classification

Neural Networks and Deep Learning Based Classification Techniques: Use H20 for Deep Learning Classification

6. H20 Deep Learning for Classification

Neural Networks and Deep Learning Based Classification Techniques: H20 Deep Learning for Classification

Course Images

Clustering and Classification with Machine Learning in Python

By Packt

Booking options

Highlights