From the course: Build GANs and Diffusion Models with TensorFlow and PyTorch
Join today to access over 24,800 courses taught by industry experts.
Transposed convolutional layer
From the course: Build GANs and Diffusion Models with TensorFlow and PyTorch
Transposed convolutional layer
- [Instructor] Before we see how the convolutional neural network architecture can be used to build generative adversarial networks, it's important to understand the different types of convolutional layers that exist. There are three types. We have the standard convolutional layer. We understand that well. Then we have deconvolutional layers. And finally, we have something known as the transposed convolutional layer. The standard convolutional layer is something that we're familiar with. We apply a sliding kernel to the input data in order to extract feature map representations of the underlying image. The deconvolutional layer is a layer that reverses the operation of a standard convolutional layer. It takes as its input a feature map and produces the original image that was used to generate that feature map. Deconvolutional layers perform the reverse operation of convolutional layers. Next, there is a third kind of…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.