Coursera – Specialization in Generative Adversarial Networks (GANs) 2021-2 – Download
Description
Specialization in Generative Adversarial Networks (GANs). is a training course for adversarial Gan networks (or GAN). GANs are powerful machine learning models that can produce realistic images, videos, and sounds. Although this model of game theory has its origins in… Today, its application is very wide, ranging from improving cybersecurity and anonymous data to data protection, creating artistic images, colorful images, black and white images, increasing resolution, etc. Create the avatar, convert a two-dimensional to a three-dimensional photo and much more. This course will introduce you to the creation of images through GAN and upgrade your knowledge from basic concepts to advanced techniques as well as for engineers, software, etc. Students and researchers from all disciplines who are interested in machine learning and understand how the GAN works are suitable.
The course is divided into 3 sections. In the first part of the. You will understand that in the root GAN upload section, a GAN is to be created simply using the module that PyTorch wants to create. The layer design is suitable for building an advanced DCGAN, capable of processing images and applying the W-loss function you use, and how to build a GAN, the condition will be familiar to you. The second part addresses the challenges of GAN evaluation, showing how the different GAN models can be compared, using FID methods to assess the truth and diversity of the models. You will be familiar with the resources for detecting bias and implementing various techniques related to StyleGAN. The final section is also dedicated to the practical use of GANs to improve data and privacy, the creation of Pix2Pix and CycleGAN for image translation and other uses.
What are you learning?
Understanding Gans components, creating simple Gans with PyTorch and advanced DCGans
Comparison of manufacturer models, use of the Fréchet Inception Distance FID method, diagnosis of ERBI and implementation of StyleGAN techniques.
Use GANs to improve privacy mapping applications and test and build Pix2Pix and CycleGAN for image translation.
What skills are you acquiring?
Generative Adversarial Networks (GANs)
Productive and interpreter from photo to photo
Controlled and conditional production
WGANs, Dcgans and StyleGANs
ERBI in the GANs
And…
Specifications of specialization in Generative Adversarial Networks (GANs).
Publisher: Coursera
Lecturers: Sharon Zhou, Eda Zhou, Eric Zelikman
Language: English
Education level: intermediate
Quantity: 3 courses
Duration of the course: With the suggested duration of 9 hours per week, approximately 3 months
Courses
- Build basic generative adversarial networks (GANs)
- Build better generative adversarial networks (GANs).
- Applying Generative Adversarial Networks (GANs)
requirements
- Learners should have practical knowledge of AI, deep learning and convolutional neural networks. You should have intermediate Python skills and some experience with a deep learning framework (TensorFlow, Keras or PyTorch). Learners should have a basic knowledge of calculus, linear algebra and statistics.
- We strongly recommend that you fill this out Specialization in deep learning before the start of GANs specialization.
Pictures
Example film
installation Guide
After extracting with the player you will get your custom view.
Subtitles: English
Quality: 720p
Download link
Download Coursera – Generative Adversarial Networks (GANs) Specialization 2021-2
Password file(s): free software
File size
635MB