philschmid blog

Machine Learning Articles

#AWS #HuggingFace #BERT #SageMaker #Serverless

Few-shot learning in practice with GPT-Neo

June 05, 2021 · 6 min read

The latest developments in NLP show that you can overcome this limitation by providing a few examples at inference time with a large language model - a technique known as Few-Shot Learning. In this blog post, we'll explain what Few-Shot Learning is, and explore how a large language model called GPT-Neo.

Distributed Training: Train BART/T5 for Summarization using 🤗 Transformers and Amazon SageMaker

April 09, 2021 · 10 min read

Learn how to train distributed models for summarization using Hugging Face Transformers and Amazon SageMaker and upload them afterwards to huggingface.co.

Multilingual Serverless XLM RoBERTa with HuggingFace, AWS Lambda

December 17, 2020 · 13 min read

Learn how to build a Multilingual Serverless BERT Question Answering API with a model size of more than 2GB and then testing it in German and France.

Serverless BERT with HuggingFace, AWS Lambda, and Docker

December 06, 2020 · 13 min read

Learn how to use the newest cutting edge computing power of AWS with the benefits of serverless architectures to leverage Google's "State-of-the-Art" NLP Model.

New Serverless Bert with Huggingface, AWS Lambda, and AWS EFS

November 15, 2020 · 15 min read

Build a serverless Question-Answering API using the Serverless Framework, AWS Lambda, AWS EFS, efsync, Terraform, the transformers Library from HuggingFace, and a `mobileBert` model from Google fine-tuned on SQuADv2.

Fine-tune a non-English GPT-2 Model with Huggingface

September 06, 2020 · 8 min read

Fine-tune non-English, German GPT-2 model with Huggingface on German recipes. Using their Trainer class and Pipeline objects.

Serverless BERT with HuggingFace and AWS Lambda

June 30, 2020 · 11 min read

Build a serverless question-answering API with BERT, HuggingFace, the Serverless Framework and AWS Lambda.

BERT Text Classification in a different language

May 22, 2020 · 8 min read

Build a non-English (German) BERT multi-class text classification model with HuggingFace and Simple Transformers.