In this video, I show you how to accelerate Transformer inference with Optimum, an open-source library by Hugging Face, and Better Transformer, a PyTorch extension available since PyTorch 1.12. Using an AWS instance equipped with an NVIDIA V100 GPU, I start from a couple of models that I previously…

In this video, I show you how to accelerate Transformer inference with Inferentia, a custom chip designed by AWS. Starting from a Hugging Face BERT model that I fine-tuned on AWS Trainium (https://youtu.be/HweP7OYNiIA), I compile it with the Neuron SDK for Inferentia. Then, using an inf1.6xlarge instance (4 Inferentia chips…

In this video, I show you how to accelerate Transformer inference with Optimum, an open source library by Hugging Face, and Intel OpenVINO. I start from a Vision Transformer model fine-tuned for image classification, and quantize it with OpenVINO. Running benchmarks on an AWS c6i instance (Intel Ice Lake architecture)…

In this video, I show you how to accelerate Transformer inference with Optimum, an open source library by Hugging Face, and ONNX. I start from a DistilBERT model fine-tuned for text classification, export it to ONNX format, then optimize it, and finally quantize it. Running benchmarks on an AWS c6i…

We have all our definition of the perfect development environment for ML. For me, it’s the combination of cloud-based managed infrastructure (all the way up to large multi-GPU instances), a great software IDE (not Jupyter!), and amazing productivity tools. In this video, I show you how to make it happen by installing VS Code and Github Copilot on Amazon SageMaker. Yes, it’s possible :) Enjoy! AWS blog: https://aws.amazon.com/blogs/machine-learning/host-code-server-on-amazon-sagemaker/

Following my test of Github Copilot, I was curious how Amazon CodeWhisperer would perform on the same example. So I gave it a try! In a nutshell, things didn’t go well. I found the service slow, and most of the prompts it generated were irrelevant. It also struggled to…

And now for something completely different! With a few hours to kill in the speaker room, I decided to take a stab at writing Hugging Face code with Github Copilot. No speech on this one, just a light rock music track, I hope you’ll enjoy it. This video was…

Building image datasets is hard work. Instead of scraping, cleaning and labeling images, why not generate them directly with a Stable Diffusion model? In this video, I show you how to generate new images with a Stable Diffusion model and the diffusers library, in order to augment an image classification…

Here are two recent webinars that you may enjoy. In the Iguazio webinar, we show you to deploy Hugging Face models with MLRun into real business environments, serve them continuously at scale, manage their lifecycle in production, and much more. In the Pinecone webinar, we show you how to…