From the course: Hands-On Introduction to PyTorch for Machine Learning
Join today to access over 25,300 courses taught by industry experts.
Understand PyTorch autograd - PyTorch Tutorial
From the course: Hands-On Introduction to PyTorch for Machine Learning
Understand PyTorch autograd
- [Instructor] In the next two lessons, we'll cover the functionalities of Autograd. First, let's have a quick refresh on Autograd and PyTorch. Autograd is PyTorch's automatic differentiation engine. It enables automatic computation of gradients, which are essential for training neural networks using back propagation. Think of it as PyTorch's internal system that keeps track of all operations on tensors and builds a dynamic computation graph. This graph is then used to compute gradients of a loss function with respect to the model's parameters. So why are gradients crucial in deep learning? In deep learning, we optimize a loss function, for example, MSE, cross-entropy, to train a neural network. To do this, we're following the three steps in the training loop. 1, compute the loss between model prediction and the true label. 2, compute gradients of the loss with respect to the model parameters. 3, update the…
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.