Real-world projects from industry experts
With real-world projects and immersive content built in partnership with top-tier companies, you’ll master the tech skills companies want.
Introduction to Generative Adversarial Networks
Course
Learn how to build and train different generative adversarial network architectures to generate new images.
Estimated time
1 month
Enroll by
May 31, 2023
Get access to classroom immediately on enrollment
Skills acquired
Generative Adversarial Networks, Hyperparameter Tuning
What You Will Learn
Building Generative Adversarial Networks
1 month to completeBecome familiar with generative adversarial networks (GANs). Students in this course will learn how to build and train different GANs architectures to generate new images. Discover, build, and train architectures such as DCGAN, CycleGAN, ProGAN, and StyleGAN on diverse datasets including the MNIST dataset, Summer2Winter Yosemite dataset, or CelebA dataset.
Prerequisite knowledge
Fully Convolutional Neural Networks, Python Proficiency, PyTorch
Generative Adversarial Networks
Build generator and discriminator using fully connected layers. Implement loss functions and train a custom GAN on the MNIST dataset.
Training a Deep Convolutional GANs
Build generator and discriminator using convolutional, batch normalization, and fully connected layers. Train a DCGAN model on the CIFAR10 dataset and implement evaluation metrics and evaluate generated samples.
Image to Image Translation
Implement unpaired data loader. Build the CycleGAN generator using residual connection and an encoder-decoder structure. Train a CycleGAN model on the summer2winter Yosemite dataset.
Modern GANs
Implement Wasserstein loss and gradient penalties and build the ProGAN generator. Implement StyleGAN components (adaptive instance normalization).
Course Project: Face Generation
Build and train a custom GAN architecture on the CelebA dataset, leveraging the different skills learned during the course. Build a custom GAN architecture, including generator and discriminator. Experiment with the different loss functions discovered during the course, such as the Binary Cross Entropy loss or the Wasserstein loss. Finally, you’ll utilize some of the methods learned to stabilize training, such as label smoothing
All Our Courses Include
Real-time support
On demand help. Receive instant help with your learning directly in the classroom. Stay on track and get unstuck.
Workspaces
Validate your understanding of concepts learned by checking the output and quality of your code in real-time.
Flexible learning program
Tailor a learning plan that fits your busy life. Learn at your own pace and reach your personal goals on the schedule that works best for you.
Course offerings
Class content
- Real-world projects
- Project reviews
- Project feedback from experienced reviewers
Student services
- Student community
- Real-time support
Succeed with personalized services.
We provide services customized for your needs at every step of your learning journey to ensure your success.
- Experienced Project Reviewers
- Real-Time Support
Get timely feedback on your projects.
- Personalized feedback
- Unlimited submissions and feedback loops
- Practical tips and industry best practices
- Additional suggested resources to improve
1,400+ project reviewers
2.7M projects reviewed
88/100 reviewer rating
1.1 hours avg project review turnaround time
Learn with the best.
Learn with the best.
Thomas Hossler
Sr Deep Learning EngineerThomas is originally a geophysicist but his passion for Computer Vision led him to become a Deep Learning engineer at various startups. By creating online courses, he is hoping to make education more accessible. When he is not coding, Thomas can be found in the mountains skiing or climbing.
Advanced Computer Vision and Deep Learning
Get started today
Learn
Learn how to build and train different generative adversarial network architectures to generate new images.
Average Time
On average, successful students take 1 month to complete this program.
Benefits include
- Real-world projects from industry experts
- Real-time support
Program Details
Do I need to apply? What are the admission criteria?
No. This Course accepts all applicants regardless of experience and specific background.
What are the prerequisites for enrollment?
A well-prepared learner has experience with Convolutional neural networks, Recurrent neural networks, Intermediate Python, PyTorch, Basic calculus, Linear algebra, Basic probability, and Jupyter notebooks.
How is this course structured?
This course is comprised of content and curriculum to support one project. We estimate that students can complete the program in one month.
The project will be reviewed by the Udacity reviewer network and platform. Feedback will be provided and if you do not pass the project, you will be asked to resubmit the project until it passes.
How long is this course?
Access to this course runs for the length of time specified in the payment card above. If you do not graduate within that time period, you will continue learning with month to month payments. See the Terms of Use and FAQs for other policies regarding the terms of access to our programs.
Can I switch my start date? Can I get a refund?
Please see the Udacity Program Terms of Use and FAQs for policies on enrollment in our programs.
What software and versions will I need in this course?
There are no software and version requirements to complete this course. All coursework and projects can be completed via Student Workspaces in the Udacity online classroom. Udacity’s full technical requirements are listed here.