Hugging Face on Amazon SageMaker
Train and deploy Hugging Face models in minutes
With Hugging Face on Amazon SageMaker, you can deploy and fine-tune pre-trained models from Hugging Face, an open-source provider of natural language processing (NLP) models known as Transformers, reducing the time it takes to set up and use these NLP models from weeks to minutes.
NLP refers to machine learning (ML) algorithms that help computers understand human language. They help with translation, intelligent search, text analysis, and more. However, NLP models can be large and complex (sometimes consisting of hundreds of millions of model parameters), and training and optimizing them requires time, resources, and skill.
AWS collaborated with Hugging Face to create Hugging Face AWS Deep Learning Containers (DLCs), which provide data scientists and ML developers a fully managed experience for building, training, and deploying state-of-the-art NLP models on Amazon SageMaker.
Get started in minutes
Hugging Face offers a library of over 10,000 Hugging Face Transformers models that you can run on Amazon SageMaker. With just a few lines of code, you can import, train, and fine-tune pre-trained NLP Transformers models such as BERT, GPT-2, RoBERTa, XLM, DistilBert, and deploy them on Amazon SageMaker.
Train and deploy at scale
Using the Amazon SageMaker distributed training libraries and Hugging Face AWS Deep Learning Containers (DLCs), you can train and deploy models in hours instead of weeks.
Generate faster predictions
Make predictions faster across instance types such as GPU, CPU, and AWS Inferentia as well as popular ML frameworks such as PyTorch and TensorFlow.
Accelerate innovation with purpose-built ML tools
Take advantage of Amazon SageMaker’s purpose-built tools for every step of the ML development lifecycle. For example, with Amazon SageMaker Pipelines, you can create, automate, and manage the end-to-end ML workflow of your NLP solution.
How it works
Hugging Face Transfomers models provide a wide variety of NLP models such as classification, information extraction, and question and answering in over 200 languages so it's easy to add sentiment analysis to your ML application. Sentiment analysis is the process of using textual data and predicting if the text expresses positive or negative sentiment.
Hugging Face provides a wide variety of models such as text processing and tokenization that make it easy to deploy text summarization in your ML application. Text summarization shortens long pieces of text, such as a news articles, into short summaries that capture the main points.
Hugging Face provides a set of predefined topics for text classification applications such as filtering spam emails and understanding customer intent based on search queries. Text classification categorizes text into a group of words. For example, you could create an ML text classification model to group news articles into topic areas such as sports, current events, and pop culture.
Blogs and articles
AWS and Hugging Face collaborate to simplify and accelerate adoption of Natural Language Processing models
March 23, 2021
Announcing managed inference for Hugging Face models in Amazon SageMaker
Sai Sharanya Nalla and Kartik Kannapur
July 8, 2021
Visit the Hugging Face documentation.
Instantly get access to the AWS Free Tier.
Get started building in the Amazon SageMaker Console.