DeepFaceLab is the most popular deepfaking software on the internet. We invited one of the contributors to teach us how to get started making deepfakes with a powerful cloud GPU machine from Paperspace.
Paperspace is one of only two notebook services recommended by Fast.ai in the 2022 edition of Practical Deep Learning for Coders from Fast.ai!
This blog details different techniques for filtering image data and explores what these filters actually do to an image as it passes through the layers of a Convolutional Neural Network (CNN). https://blog.paperspace.com/filters-in-convolutional-neural-networks/
In this tutorial, we examine the DALL-E family of image generation frameworks from OpenAI.
In this tutorial, we examine mixed-precision training to try and understand how we can leverage it in our code, how it fits into the traditional DL algorithmic paradigm, what frameworks support mixed precision training, and performance tips on using GPUs for DL with automatic mixed precision.
This deep learning tutorial overview covers mixed precision training, the hardware required to take advantage of such computational capability, and the advantages of using mixed precision training in detail.
In this article, we break down the justification and inspiration for DALL-E Mini/Craiyon, explore its predecessors for comparison's sake, and implement the light image generator in Python code.
In this article, we will see why GANs are awesome, understand what GANs really are, how they work, dive deep into the loss function that they use, and then build our first simple GAN from scratch to generate MNIST.
In this article, we explore the progress that deep learning has made in the field of music in numerous tasks related to audio and signal processing. We then proceed to model and generate our own music files using pretty_midi.
