From the course: Rust for Data Engineering
Unlock this course with a free trial
Join today to access over 24,900 courses taught by industry experts.
Rust GPU Hugging Face translator - Rust Tutorial
From the course: Rust for Data Engineering
Rust GPU Hugging Face translator
- [Instructor] One of the most powerful ways to use Rust is to dive into using things like pre-trained models, using GPU, building high performance command line tools that you can do batch-based machine learning inference against. What I'm going to show you is how simple it is to use GitHub Codespaces with A GPU, enable that GPU, build up the Rust bindings for it and take a translation tool and build it out in just a few minutes. Okay, let's go ahead and get started. Here I've got a repo nogibjj, rust-pytorch-gpu-template that has all the things you need to get GPUs cooking in Codespaces. And what I'm going to take a look at here is the diagram of what we're going to build. So to start with, you have a Cuda GPU enabled in Codespaces and then I'm going to go through and use the Rust bindings for Hugging Face. Again, Rust is a high performance language that allows you to build very performant tools and then we're going to…
Contents
-
-
-
-
-
-
-
-
-
-
-
-
Introduction to Hugging Face Hub5m 9s
-
(Locked)
Rust PyTorch pre-trained model ecosystem3m 39s
-
(Locked)
Rust GPU Hugging Face translator6m 14s
-
(Locked)
Rust PyTorch high-performance options7m 52s
-
(Locked)
EFS ONNX Rust inference with AWS Lambda9m 38s
-
(Locked)
Theory behind model fine-tuning2m 54s
-
(Locked)
Doing fine-tuning8m 9s
-
-