AI
Using & Mixing Hugging Face Models with Gradio 2.0
Hugging Face has released an update to its Gradio library, which allows users to mix and match different pre-trained models from their model hub. This...
AI
AI
OpenAI Scholars 2021: Final projects
The OpenAI Scholars 2021 program has concluded, with participants completing a six-month mentorship and producing open-source research projects. Each ...
AI
Will Hurd joins OpenAI’s board of directors
OpenAI has added Congressman Will Hurd to its board of directors. The company believes that developing general-purpose AI requires expertise in both p...
AI
Scaling-up BERT Inference on CPU (Part 1)
Hugging Face explores ways to speed up the inference process of BERT, a popular AI model, on a standard CPU. They discuss various techniques for scali...
AI
Introducing 🤗 Accelerate
Hugging Face has introduced a new library called 🤗 Accelerate, which is designed to simplify the process of training and deploying large language mode...
AI
Distributed Training: Train BART/T5 for Summarization using 🤗 Transformers and Amazon SageMaker
Researchers have developed a method to train large-scale language models like BART and T5 using distributed training on Amazon SageMaker. This approac...
AI
Understanding BigBird's Block Sparse Attention
Researchers have introduced a new attention mechanism called Block Sparse Attention, which is part of the BigBird model. This mechanism allows for mor...
AI
GPT-3 powers the next generation of apps
Over 300 apps have integrated OpenAI's GPT-3 API to offer users advanced AI capabilities such as search, conversation, and text completion. This integ...
AI
The Partnership: Amazon SageMaker and Hugging Face
Amazon SageMaker, a cloud-based machine learning platform, has partnered with Hugging Face, an open-source AI model library. The partnership will allo...
AI
Reducing Toxicity in Language Models
Large pretrained language models, trained on vast amounts of online data, often acquire toxic behavior and biases from the internet. This is a concern...
AI
My Journey to a serverless transformers pipeline on Google Cloud
A developer describes their process for deploying a Transformers pipeline on Google Cloud's serverless infrastructure. They use the Hugging Face libra...
AI
Fine-Tune Wav2Vec2 for English ASR in Hugging Face with 🤗 Transformers
Hugging Face has released a tutorial on fine-tuning the Wav2Vec2 model for English automatic speech recognition (ASR) tasks. The tutorial uses Hugging...