From the course: Hands-On Introduction to Transformers for Computer Vision
Join today to access over 24,700 courses taught by industry experts.
Fine-tuning Swin Transformers - PyTorch Tutorial
From the course: Hands-On Introduction to Transformers for Computer Vision
Fine-tuning Swin Transformers
- [Instructor] Hey, everyone. Welcome back to Fine-Tuning a Swin transformer on CIFAR-10. This time, we are going to be leveling up. We're not going to be building it from scratch. Instead, we're going to be using one of the best available today, these Swin Transformer to train on a very similar dataset to MNIST, but a little bit bigger. Now we're doing CIFAR-10, these aren't pixelized numbers, these are now pixelized little animals and things like that. So first and foremost, (chuckling) this one is probably our most important one, that if you are trying to run this locally, versus if you're trying to run this in a Codespace or a CoLab or wherever you might be running this online, it might take very long. Don't be discouraged, this is just the reality of machine learning, that the better models you want to train, especially when it comes to transformers, are more computationally expensive. So if you have a GPU, great, you should definitely be using it. If you don't, or even if you do…
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.