Teach on Udemy

Turn what you know into an opportunity and reach millions around the world.

Learn More

Your cart is empty.

Keep shopping

Introduction to Machine Learning

Name: Introduction to Machine Learning
Rating: 4.5 (18 reviews)

Linear and Logistic Regression and Neural Networks Using Python

Created bySatish Reddy

Last updated 9/2020

English

What you'll learn

Introduction to Machine Learning: math, algorithms, and Python coding for Linear and Logistic Regression and Neural Networks

Course content

9 sections • 120 lectures • 14h 31m total length

Section 1.1: What is Machine Learning?5:30
Overview of Machine Learning and areas of application
Section 1.2: Linear and Logistic Regression and Neural Networks12:56
Examples of Supervised Machine Learning and introduction to linear regression, logistic regression, and neural network approaches to be discussed in this course.
Section 1.3: About this Course5:49
Info about the course: the prerequisites, how best to learn from the material, course sections and information about the instructor.
Section 1.4: Resources and Set Up8:47
Resources and set up (environment creation and Python package installation) for the course.
Section 1.4: DEMO11:10
Demo of resources and set up of environment and packages and Jupyter notebook test.

Section 2.0: Python Demos1:05
Overview of Python demos in Section 2.
Section 2.1: Numpy1:49
Overview of numpy functionality used in this course.
Section 2.1: DEMO13:31
Jupyter notebook demo of numpy functionality used in the course
Section 2.1: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 2.2: Matplotlib Basics1:14
Overview of matplotlib functionality used in this course.
Section 2.2: DEMO11:26
Jupyter notebook demo of matplotlib functionality used in the course.
Section 2.2: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 2.3: Matplotlib Heatmap3:11
This section gives an overview of matplotlib heatmap creation functionality used in this course.
Section 2.3: DEMO11:23
Jupyter notebook demo of matplotlib heatmap creation functionality used in the course.
Section 2.3: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 2.4: Pandas1:39
Overview of pandas functionality used in this course.
Section 2.4: DEMO7:31
Jupyter notebook demo of pandas functionality used in the course.
Section 2.4: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 2.5: unittest2:06
Overview of unittest functionality used in this course.
Section 2.5: DEMO4:01
Jupyter notebook demo of unittest functionality used in the course.
Section 2.5: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 2.6: sklearn3:40
Overview of sklearn (scikit-learn) functionality used in this course.
Section 2.6: DEMO7:18
Jupyter notebook demo of sklearn functionality used in the course.
Section 2.6: Exercises0:35
Exercises Section - see pdf file in Resources link

Section 3.0: Review of Mathematical Concepts2:43
Overview of Section 3, which covers linear algebra, multi-variable calculus, and optimization
Section 3.1: Linear Algebra14:38
Brief review of matrix notation, transposes, multiplication, and broadcasting
Section 3.1: DEMO5:17
Jupyter notebook demo about numpy functionality for transposes, multiplication, and broadcasting.
Section 3.1: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 3.2: Multi-variable Calculus20:03
Review of partial derivatives, gradients and the chain rule of multi-variable calculus
Section 3.2: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 3.3: Optimization12:15
Gradient Descent algorithm for minimizing a function
Section 3.3: DEMO7:37
Jupyter notebook demo of the Gradient Descent algorithm.
Section 3.3: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 3.4: Notation1:12
Notation used in course.

Section 4.0: Linear and Logistic Regression Overview1:49
Overview of section on Linear and Logistic Regression
Section 4.1: Linear Regression: Normal Equations4:37
Review of normal equations approach for linear regression.
Section 4.1: DEMO7:03
Jupyter notebook demo of normal equations.
Section 4.2: Linear Regression: Mathematical Foundations27:44
Mathematical foundations for linear regression, including forward and back propagation, loss, training, prediction, and accuracy calculation.
Section 4.2: DEMO8:57
Jupyter notebook demo of the machine learning approach for linear regression.
Section 4.2: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 4.3: Derivative Testing11:26
Approach for testing the forward/back propagation algorithm for computing derivatives.
Section 4.3: DEMO5:48
Jupyter notebook demo of derivative testing.
Section 4.3: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 4.4: Code Overview5:39
Broad overview of course code framework design.
Section 4.5: Walkthrough Version 1.18:18
Description of codes in Version1.1 of framework, which implements forward and back propagation and derivative unit test.
Section 4.5: DEMO16:35
Walkthrough of Version1.1 of framework.
Section 4.5: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 4.6: Walkthrough Version 1.26:52
Description of Version1.2 of the framework, which includes codes for training, prediction, accuracy and driver for linear regression.
Section 4.6: DEMO20:11
Walkthrough of Version1.2 of framework.
Section 4.6: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 4.7: Logistic Regression: Mathematical Foundations23:21
Mathematical foundations for logistic regression
Section 4.7: DEMO7:46
Jupyter notebook demo of logistic regression calculations.
Section 4.7: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 4.8: Walkthrough Version1.35:39
Descriptions of Version1.3 of the framework, which adds loss, activation, unit test, and driver for logistic regression.
Section 4.8: DEMO22:14
Walkthrough of Version1.3 of the framework.
Section 4.8: Exercises0:35
Exercises Section - see pdf file in Resources link

Section 5.0: Neural Networks Overview1:38
Overview of the Neural Networks Section 5.0.
Section 5.1: Neural Networks: Mathematical Foundations27:29
Mathematical foundations for neural networks, including forward and back propagation, training, prediction, and accuracy calculation.
Section 5.1: DEMO7:38
Jupyter notebook demo for neural networks
Section 5.1: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 5.2: Implementation of Activation Functions5:52
Implementation of activation functions to avoid overflow issues.
Section 5.2: DEMO8:20
Jupyter notebook demo of implementation of activation functions.
Section 5.2: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 5.3: Walkthrough Version 2.16:23
Description of Version2.1 of the framework, which includes implementation of neural networks for binary classification.
Section 5.3: DEMO23:31
Walkthrough of Version2.1 of the framework.
Section 5.3: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 5.4: Softmax Activation Function13:29
Definition and practical implementation of softmax activation function used for multiclass classification.
Section 5.4: DEMO5:21
This section is Jupyter notebook demo for the softmax activation function.
Section 5.4: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 5.5: One-Hot Matrix4:37
One-hot matrix and its inverse.
Section 5.5: DEMO4:58
Jupyter notebook demo of the one-hot matrix and its inverse.
Section 5.5: Exercises0:35
Exercises Section - see pdf file in Resources link
Section 5.6: Multiclass Classification: Mathematical Foundations26:54
Mathematical foundations for neural networks for multiclass classification, including forward and back propagation, prediction, and accuracy calculation.
Section 5.6: DEMO9:06
Jupyter notebook demo of neural networks for multiclass classification.
Section 5.6: Exercises0:35
Exercises section - see pdf file in Resources link.
Section 5.7: Walkthrough Version 2.23:36
Description of Version2.2 of the code, which implements of multiclass classification
Section 5.7: DEMO16:14
Walkthrough of Version2.1 of the framework.
Section 5.7: Exercises0:35
Exercises section - see pdf file in Resources link.

Section 6.0: Overview1:33
Overview of Section 6 on Optimization, Validation, Accuracy, and Regularization.
Section 6.1: Stochastic and Min-Batch Optimization20:26
Description of stochastic and mini-batch optimization.
Section 6.1: DEMO8:38
Jupyter notebook demo of stochastic and mini-batch optimization.
Section 6.1: Exercises0:35
Exercises section - see pdf file in Resources link.
Section 6.2: Momentum, RmsProp, Adam Optimization15:39
Description and examples of momentum, RmsProp, and Adam optimization
Section 6.2: DEMO7:05
Jupyter notebook demo of momentum, RmsProp, and Adam optimization.
Section 6.2: Exercises0:35
Exercises section - see PDF file in Resources link.
Section 6.3: Walkthrough Version 3.14:05
Description of framework Version3.1, which includes implementation of mini-batch, momentum, RmsProp, and Adam optimization.
Section 6.3: DEMO16:52
Walkthrough of Version 3.1 of the framework.
Section 6.3: Exercises0:35
Exercises section - see pdf file in Resources link.
Section 6.4: Validation8:34
Presentations of approaches for validation of machine learning systems.
Section 6.5: Performance Measures10:39
Description of alternatives to accuracy, such as precision, recall, f1score, and confusion matrix, for measuring performance.
Section 6.5: Exercises0:35
Exercise section - see PDF file in Resources link.
Section 6.6: Walkthrough Version 3.23:58
Description of Version3.2 of framework, which includes new validation and performance measures.
Section 6.6: DEMO21:59
Walkthough of Version3.2 of framework.
Section 6.6: Exercises0:35
Exercises section - see PDF file in Resources link.
Section 6.7: Regularization10:36
Description of regularization for machine learning systems.
Section 6.8: Hyperparameter Search6:06
Description of hyperparameter search process for tuning a machine learning system.
Section 6.9: Walkthrough Version 3.33:32
Description of framework Version 3.3, which includes implementation of regularization and hyperparameter search.
Section 6.9: DEMO16:55
Walkthrough of Version3.3 of framework.
Section 6.9: Exercises0:35
Exercises section - see pdf file in Resources link.
Section 6.10: Underfitting and Overfitting22:10
Description of diagnosis and remediation of underfitting and overfitting.
6.10: Exercises0:35
Exercises section - see PDF file in Resources link.

Section 7.0: Case Studies1:37
Overview of the Section 7, which covers 3 case studies.
Section 7.1: House Price Prediction11:38
Description of approach of using linear regression to predict house prices.
Section 7.1: DEMO11:02
Walkthrough of house prediction case study.
Section 7.1: Exercises0:35
Exercises Section - see PDF file in Resources link.
Section 7.2: Spam Classification13:47
Description of approach for using neural network spam classification.
Section 7.2: DEMO15:00
Walkthrough of spam classification case study.
Section 7.2: Exercises0:35
Exercises section - see PDF file in Resources Link.
Section 7.3: Case Study: MNIST Digits Classification16:19
Description of approach for using neural networks for digits classification.
Section 7.3: DEMO17:00
Walkthrough of digits classification case study.
Section 7.3: Exercises0:35
Exercises section - see PDF file in Resources link.

Section 8.0: Introduction to Tensorflow2:24
Overview of the Tensorflow Section.
Section 8.1: Tensorflow for MNIST Digits Classification13:04
Overview of Tensorflow code for MNIST digits classification and comparison with course framework.
Section 8.1: DEMO9:29
Walkthrough of use Tensorflow for the MNIST digits classification.
Section 8.1: Exercises0:35
Exercises section - PDF file in Resources link.
Section 8.2: Demo of Tensorflow on a GPU12:01
Discussion of GPU and walkthrough use of Tensorflow for MNIST digits classification on a GPU.

Section 9.0: Summary and Thank You0:42
Overview of the Summary and Thank You section.
Section 9.1: Alternative Approaches6:01
This section has a brief overview of alternative approaches, including decision trees, random forest, and Adaboost.
Section 9.2: Course Summary11:19
This section has a summary of the course.
Section 9.3: Suggestions for Further Study and Thank You4:52
Listing of sources for further study and provides a thank you for students.

Requirements

Linear Algebra (matrix multiplication), Multivariable Calculus (gradients), Python programming

Description

Course Outcome:

After taking this course, students will be able to understand and implement machine learning algorithms using Python for regression, binary classification, and multi-class classification with applications to real-world datasets.

Course Topics and Approach:

This introductory course on machine learning focuses on Supervised Learning, which involves finding functions that fit data and then using the functions to make predictions. Applications include image classification, text sentiment classification, house price prediction.

The core of this course involves study of the following algorithms:

Linear Regression, Logistic Regression, Neural Networks for regression, binary, and multiclass classification

Unlike many other courses, this course:

Has a detailed presentation of the the math underlying the above algorithms including optimization algorithms and back propagation formulas
Has a detailed explanation of how algorithms are converted into Python code with lectures on code design and use of vectorization
Has homework questions (programming and theory) and solutions that allow learners to get practice with the course material

The course codes are then used to address case studies involving real-world data including image classification, text message spam/no spam classification, and house price prediction.

Course Audience:

This course is designed for:

Scientists, engineers, and programmers and others interested in machine learning/data science
No prior experience with machine learning is needed
Students should have knowledge of
- Basic linear algebra (vectors, matrix multiplication, transpose)
- Multivariable calculus (to follow details of optimization and backpropagation formulas)
- Python 3 programming

Students should have a Python installation, such as the Anaconda platform, on their machine with the ability to run programs in the command window and in Jupyter Notebooks

Teaching Style and Resources:

Course includes many examples with plots used to help students get a better understanding of the material
Course has 50+ exercises with solutions (theoretical, Jupyter Notebook, and programming) to allow students to gain additional practice
All resources (presentations, supplementary documents, demos, codes, solutions to exercises) are downloadable from the course Github site.

Who this course is for:

Students without any previous experience with machine learning
Students with previous experience who would like to revisit the math, algorithms and coding for machine learning in detail

Introduction to Machine Learning

What you'll learn

Explore related topics

Course content

Introduction5 lectures • 44min

Python Demos19 lectures • 1hr 13min

Review of Mathematical Concepts10 lectures • 1hr 6min

Linear and Logistic Regression22 lectures • 3hr 7min

Neural Networks22 lectures • 2hr 49min

Optimization, Validation, Accuracy & Regularization23 lectures • 3hr 3min

Case Studies10 lectures • 1hr 28min

Introduction to Tensorflow5 lectures • 38min

Summary and Thank You4 lectures • 23min

Requirements

Description

Who this course is for: