Explore machine learning's essentials for software engineers, delve into supervised learning, neural networks, and deep learning, and gain skills to tackle real-world data challenges effectively.

pml_final.tar.gz

just_run

chapter06

SPA_07

SPA_gdb_frontEnd

J12-1

J12-2

SPA-server_image

Handson-10

Machine learning is the future for the next generation of software professionals.

This course serves as a guide to machine learning for software engineers. You’ll be introduced to three of the most relevant components of the AI/ML discipline; supervised learning, neural networks, and deep learning. You’ll grasp the differences between traditional programming and machine learning by hands-on development in supervised learning before building out complex distributed applications with neural networks. You’ll go even further by layering networks to create deep learning systems. You’ll work with complex real-world datasets to explore machine behavior from scratch at each phase.

By the end of this course, you’ll have a working knowledge of modern machine learning techniques. Using software engineering, you’ll be prepared to build complex neural networks and wrangle real-world data challenges.

Fundamentals of Machine Learning for Software Engineers

Which activation function do we prefer at the output layer of a non-binary output (or categorical output)?


Which activation function do we prefer at the output layer of a non-binary output (or categorical output)?

A sigmoid gets saturated due to large input, causing the sigmoid’s gradient to tend to zero.

Some of the nodes in the hidden layer vanish due to close to zero inputs.

Sigmoid diminishes the total gradient of the network such that the first layer in the network gets a zero gradient.

What is the main reason for dead neurons appearing in the network?

What is the main reason for dead neurons appearing in the network?

**(Fill in the blank.)** Leaky ReLU helps to avoid ________.


(Fill in the blank.) Leaky ReLU helps to avoid ________.

**(Fill in the blank.)** Without an activation function, the neural network collapses to ________.


(Fill in the blank.) Without an activation function, the neural network collapses to ________.

Making decisions based on the cross entropy loss

Identity function as activation function in the output layer

**( Select all that apply.)** If a network has numerical output, which would be true?


( Select all that apply.) If a network has numerical output, which would be true?

By randomly turning off some nodes in the network

By dropping certain layers in the network

How do we use _**dropout**_ to combat overfitting?

How do we use dropout to combat overfitting?

Test yourself on the concepts of taming deep networks by taking the quiz.learning.

How Machine Learning Works

Our First Learning Program

Walking the Gradient

Hyperspace

A Discern Machine

Get Real

The Final Challenge

The Perceptron

Designing the Network

Building the Network

Training the Network

How Classifiers Work

Batchin’ Up

The Zen of Testing

Let’s Do Development

A Deeper Kind of Network

Defeating Overfitting

Taming Deep Networks

Beyond Vanilla Networks

Into the Deep

Quiz: Tame Deep Networks