Batch normalization is a term commonly mentioned in the context of convolutional neural networks. In this article, we are going to explore what it actually entails and its effects, if any, on the performance or overall behavior of convolutional neural networks.
In this blog post, we examine Captum, which supplies academics and developers with cutting-edge techniques, such as Integrated Gradients, that make it simple to identify the elements that contribute to a model's output. We then put these techniques to use in a coding demo with ResNet.
In this article, we examine the effects of batch size on DL model training times and accuracies, and go on to describe a methodology for finding the maximum batch size for any given task.
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 tutorial, we examine the Few-Shot Learning paradigm for deep and machine learning tasks. Readers can expect to learn what it is, different techniques, and details about use cases for Few-Shot Learning
In this overview, we look at AMPT-GA: a system that selects application-level data precisions to maximize performance while satisfying accuracy constraints via automatic mixed precision training.
This guide breaks down the capabilities of the Tensor Core technology used by the latest generations of Nvidia GPUs.
This blog post details the concept of mixed precision training, its benefits, and how to implement it automatically with popular Deep Learning frameworks PyTorch and TensorFlow.
