Generative Deep Learning with TensorFlow from Coursera

In this course, you will:

a) Learn neural style transfer using transfer learning: extract the content of an image (eg. swan), and the style of a painting (eg. cubist or impressionist), and combine the content and style into a new image.

b) Build simple AutoEncoders on the familiar MNIST dataset, and more complex deep and convolutional architectures on the Fashion MNIST dataset, understand the difference in results of the DNN and CNN AutoEncoder models, identify ways to de-noise noisy images, and build a CNN AutoEncoder using TensorFlow to output a clean image from a noisy one.

c) Explore Variational AutoEncoders (VAEs) to generate entirely new data, and generate anime faces to compare them against reference images.

d) Learn about GANs; their invention, properties, architecture, and how they vary from VAEs, understand the function of the generator and the discriminator within the model, the concept of 2 training phases and the role of introduced noise, and build your own GAN that can generate faces.

The DeepLearning.AI TensorFlow: Advanced Techniques Specialization introduces the features of TensorFlow that provide learners with more control over their model architecture, and gives them the tools to create and train advanced ML models.

This Specialization is for early and mid-career software and machine learning engineers with a foundational understanding of TensorFlow who are looking to expand their knowledge and skill set by learning advanced TensorFlow features to build powerful models.

What's inside

Syllabus

Week 1: Style Transfer

This week, you will learn how to extract the content of an image (such as a swan), and the style of a painting (such as cubist, or impressionist), and combine the content and style into a new image. This is called neural style transfer, and you'll learn how to extract these kinds of features using transfer learning.

Week 2: AutoEncoders

This week, you’ll get an overview of AutoEncoders and how to build them with TensorFlow. You'll learn how to build a simple AutoEncoder on the familiar MNIST dataset, before diving into more complicated deep and convolutional architectures that you'll build on the Fashion MNIST dataset. You'll get to see the difference in results of the DNN and CNN AutoEncoder models, and then identify ways to denoise noisy images. You'll finish the week building a CNN AutoEncoder using TensorFlow to output a clean image from a noisy one!

Week 3: Variational AutoEncoders

This week you will explore Variational AutoEncoders (VAEs) to generate entirely new data. In this week’s assignment, you will generate anime faces and compare them against reference images.

Week 4: GANs

This week, you’ll learn about GANs. You'll learn what they are, who invented them, their architecture and how they vary from VAEs. You'll get to see the function of the generator and the discriminator within the model, and the concept of 2 training phases and the role of introduced noise. Then you'll end the week building your own GAN that can generate faces! How cool is that!

Good to know

Know what's good

, what to watch for

, and possible dealbreakers

Builds a strong foundation for beginners and strengthens an existing foundation for intermediate learners in Deep Learning

Develops professional and specialized skills in Advanced Techniques of Deep Learning

Taught by Laurence Moroney and Eddy Shyu, recognized experts in AI and Machine Learning

Uses hands-on labs and interactive materials for practical implementation of concepts

Requires foundational understanding of TensorFlow, limiting its accessibility to beginners

Course materials may require access to additional software or resources not readily available in all settings

Reviews summary

Tensorflow generative learning

Learners say this course in TensorFlow is well received, engaging, helpful, and challenging. The code is especially valuable in helping students understand the content. Quizzes and labs are challenging but give an opportunity to apply what they have learned. The course covers advanced topics like GANs, autoencoders, VAEs, and image processing. Students appreciate the expertise of the instructors and the well-explained content.

Covers advanced topics in deep learning.

"Really good content covering the surface of lot of advanced topics."

"Outstanding course that deals with complex topics in Deep Learning explained in short yet precise manner and flawlessly executed."

"Excellent! Outstanding!"

Engaging assignments and labs.

"Excellent course.I really appreciated to have a quiz and an assignment each week.Thanks to all the contributors."

"The quizzes and labs are quite challenging."

"sessions on KL divergence, reconstruction loss would have helped learners a lot."

Helpful code examples.

"the code really help me deeply understand these methods"

"really great course, it showed how VAE and AutoEncoders work, also touched on the topic of GANs"

"the best part was applying what's learned during the whole specialization on building difficult and complicated models from scratch."

Expert instruction by top developer.

"Excellent course - Indepth knowledge delivered by one of the top-developers in an engaginand challenging manner."

"Laurence and DeepLearning.ai team did great job."

"Sessions, labs and assignment are really very good from advance programming in Tensorflow perspective."

Focuses on image processing.

"After the whole specialization you can't say that it didn't give you an opportunity to learn how to use Tensorflow. However, it's focused mostly on image processing so if you dislike this topic - it's not for you."

Activities

Be better prepared before your course. Deepen your understanding during and after it. Supplement your coursework and achieve mastery of the topics covered in Generative Deep Learning with TensorFlow with these activities:

TensorFlow Meetup

Show steps

Expand your network and learn from others by attending a TensorFlow meetup.

Browse courses on TensorFlow

Show steps

Find a TensorFlow meetup in your area
Attend the meetup and participate in the discussions

TensorFlow Review

Show steps

Review the basics of TensorFlow before the start of this course.

Browse courses on TensorFlow

Show steps

Re-read the TensorFlow tutorial
Review the TensorFlow documentation

TensorFlow Mentor

Show steps

Help others learn TensorFlow by becoming a mentor.

Browse courses on TensorFlow

Show steps

Join a TensorFlow community or forum
Answer questions and provide guidance to other TensorFlow users

Six other activities

Expand to see all activities and additional details

Show all nine activities

TensorFlow Practice

Show steps

Practice using TensorFlow before the start of this course.

Browse courses on TensorFlow

Show steps

Complete the TensorFlow exercises on Kaggle
Create a simple TensorFlow model

TensorFlow Tutorial for Beginners

Show steps

Complete this tutorial to gain familiarity with the TensorFlow framework before the start of this course.

Browse courses on TensorFlow

Show steps

Read the tensorflow.org tutorial
Complete the exercises in the tutorial
Install TensorFlow on your computer
Create a simple TensorFlow model

TensorFlow Workshop

Show steps

Deepen your understanding of TensorFlow by attending a workshop.

Browse courses on TensorFlow

Show steps

Find a TensorFlow workshop in your area
Attend the workshop and participate in the exercises

TensorFlow Exercises

Show steps

Complete these exercises to reinforce your understanding of TensorFlow.

Browse courses on TensorFlow

Show steps

Solve the TensorFlow exercises on Kaggle
Create a TensorFlow model to solve a problem of your choice

TensorFlow Blog Post

Show steps

Share your knowledge of TensorFlow by writing a blog post.

Browse courses on TensorFlow

Show steps

Choose a topic related to TensorFlow
Research the topic and write a blog post that explains it clearly and concisely
Post your blog post on a platform of your choice

TensorFlow Project

Show steps

Complete this project to demonstrate your mastery of TensorFlow.

Browse courses on TensorFlow

Show steps

Identify a problem that can be solved using TensorFlow
Design and implement a TensorFlow model to solve the problem
Evaluate the performance of your model
Write a report summarizing your project

Career center

Learners who complete Generative Deep Learning with TensorFlow will develop knowledge and skills that may be useful to these careers:

Artificial Intelligence Engineer

Artificial Intelligence Engineers design, develop, and implement artificial intelligence systems. They also work with users to gather requirements and design AI systems that meet their needs. This course may be useful for Artificial Intelligence Engineers because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of artificial intelligence and can help Artificial Intelligence Engineers develop new and innovative AI systems.

See salaries and explore the career path for Artificial Intelligence Engineer

Computational Biologist

Computational Biologists use computational methods to analyze biological data. They also develop and implement computational models to simulate biological systems. This course may be useful for Computational Biologists because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of computational biology and can help Computational Biologists develop new and innovative computational models to simulate biological systems.

See salaries and explore the career path for Computational Biologist

Machine Learning Engineer

Machine Learning Engineers design, develop, and implement machine learning algorithms and models. They also work with data scientists to collect and prepare data for machine learning models. This course may be useful for Machine Learning Engineers because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of machine learning and can help Machine Learning Engineers develop new and innovative machine learning models.

See salaries and explore the career path for Machine Learning Engineer

Data Analyst

Data Analysts use data to make informed decisions. They also develop and implement data models to predict future outcomes. This course may be useful for Data Analysts because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of data analysis and can help Data Analysts develop new and innovative data models to predict future outcomes.

See salaries and explore the career path for Data Analyst

Computer Vision Engineer

Computer Vision Engineers design, develop, and implement computer vision systems. They also work with users to gather requirements and design computer vision systems that meet their needs. This course may be useful for Computer Vision Engineers because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of computer vision and can help Computer Vision Engineers develop new and innovative computer vision systems.

See salaries and explore the career path for Computer Vision Engineer

Robotics Engineer

Robotics Engineers design, develop, and implement robotics systems. They also work with users to gather requirements and design robotics systems that meet their needs. This course may be useful for Robotics Engineers because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of robotics and can help Robotics Engineers develop new and innovative robotics systems.

See salaries and explore the career path for Robotics Engineer

Natural Language Processing Engineer

Natural Language Processing Engineers design, develop, and implement natural language processing systems. They also work with users to gather requirements and design NLP systems that meet their needs. This course may be useful for Natural Language Processing Engineers because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of natural language processing and can help Natural Language Processing Engineers develop new and innovative NLP systems.

See salaries and explore the career path for Natural Language Processing Engineer

Data Scientist

Data Scientists collect, analyze, and interpret data to help businesses make informed decisions. They also develop and implement machine learning models to automate tasks and improve decision-making. This course may be useful for Data Scientists because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of data science and can help Data Scientists develop new and innovative data science solutions.

See salaries and explore the career path for Data Scientist

Bioinformatics Engineer

Bioinformatics Engineers design, develop, and implement bioinformatics systems. They also work with users to gather requirements and design bioinformatics systems that meet their needs. This course may be useful for Bioinformatics Engineers because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of bioinformatics and can help Bioinformatics Engineers develop new and innovative bioinformatics systems.

See salaries and explore the career path for Bioinformatics Engineer

Statistician

Statisticians use mathematical and statistical methods to collect, analyze, and interpret data. They also develop and implement statistical models to predict future outcomes. This course may be useful for Statisticians because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of statistics and can help Statisticians develop new and innovative statistical models to predict future outcomes.

See salaries and explore the career path for Statistician

Quantitative Analyst

Quantitative Analysts use mathematical and statistical methods to analyze financial data. They also develop and implement quantitative models to predict financial market trends. This course may be useful for Quantitative Analysts because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of quantitative finance and can help Quantitative Analysts develop new and innovative quantitative models to predict financial market trends.

See salaries and explore the career path for Quantitative Analyst

Actuary

Actuaries use mathematical and statistical methods to assess risk and uncertainty. They also develop and implement actuarial models to predict the probability of future events. This course may be useful for Actuaries because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of actuarial science and can help Actuaries develop new and innovative actuarial models to predict the probability of future events.

See salaries and explore the career path for Actuary

Financial Analyst

Financial Analysts use financial data to analyze and make recommendations on investments. They also develop and implement financial models to predict financial market trends. This course may be useful for Financial Analysts because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of finance and can help Financial Analysts develop new and innovative financial models to predict financial market trends.

See salaries and explore the career path for Financial Analyst

Software Engineer

Software Engineers design, develop, and implement software applications. They also work with users to gather requirements and design software that meets their needs. This course may be useful for Software Engineers because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of software engineering and can help Software Engineers develop new and innovative software applications.

See salaries and explore the career path for Software Engineer

Computer and Information Research Scientist

Computer and Information Research Scientists research, design, develop, and test computer systems and applications. They also analyze user needs and develop solutions to computing problems. This course may be useful for Computer and Information Research Scientists because it covers topics such as neural style transfer, autoencoders, and generative adversarial networks (GANs). These topics are all relevant to the field of computer science and can help Computer and Information Research Scientists develop new and innovative computing solutions.

See salaries and explore the career path for Computer and Information Research Scientist