|
Listen to this story
|
“NLP is going to be the most transformational tech of the decade!” Clément Delangue, a co-founder of Hugging Face, tweeted in 2020 – and his brainchild will definitely be remembered as a pioneer in this game-changing technology.
Giving itself, “The AI community building the future” tag, Hugging Face has emerged as one of the most influential names in the NLP technology domain. Hugging Face was founded by Clément Delangue and Julien Chaumond in 2016 as a chatbot company. Not even a decade old, the company has already roped in big names in its kitty like Apple, Monzo, Bing, and Facebook (now Meta) to use its offerings.
What does it offer?
The products offered by Hugging Face can be divided into two categories- open-source platform and subscription-based features that NLP practitioners can deploy in their models. Hugging Face provides models for a variety of tasks. Some of them are:
In NLP:
- Summarisation, question answering, table question answering, text classification, fill mask
- Sentence similarity, translation, token classification, feature extraction
In Audio:
- Text to speech
- Audio-to-audio, audio classification, voice activity detection, automatic speech recognition
- Image classification, object detection
- Image segmentation, text to image
So popular, but why?
Hugging Face has seen rapid growth in its popularity since the get-go. It is definitely doing the right things to attract more and more people to its platform, some of which are on the following lines:
- Community driven approach through large open source repositories along with paid services. Helps to build a network of like-minded people passionate about open source.
- Attractive price point. The subscription-based features, e.g.: Inference based API, starts at a price of $9/month.
- Easy interface to work with uncomplicated deployment
- Fun and playful branding in the complex world of machine learning and NLP to attract people
- Push towards ethical AI
GitHub of Machine Learning
In a media interview in March, Hugging Face CTO Julien Chaumond had said that the democratisation of AI will be one of the biggest achievements for society. He added that no single company, not even a Big Tech business, can do it alone. The company strongly believes in this vision. It has been proactively working to build the open-source community for the development of language models.
The Transformers library is one of the most popular attractions Hugging Face offers. It is backed by deep learning libraries– PyTorch and TensorFlow. In 2021, the company stepped up its game and released a variety of open-source datasets and models, making its community power even more strong.
Stepped up its open-source efforts in 2021
The company released an open-source library called Optimum, which is an optimisation toolkit for transformers at scale. This toolkit also enables maximum efficiency to train and run models on specific hardware. A researcher from Avignon University recently released an open-source, easy-to-use wrapper to Hugging Face for Healthcare Computer Vision, called HugsVision. It will find applications in image classification, semantic segmentation, object detection, and image generation.
It also released Datasets, a community library for contemporary NLP. The Datasets library contains 650 unique datasets and has more than 250 contributors. To attract the student community into understanding NLP and its applications, the company also launched the first part of its NLP course in June this year.
Ethics and AI
Margaret Mitchell, former co-head of Google’s Ethical AI research group, has joined Hugging Face to create tools that help to build algorithms that are fair. She was under the limelight after having been fired from Google, as per reports, as the aftermath of a controversy over a critical paper she had co-written.
Removing Bias from Algorithms
At Hugging Face, Mitchell’s work will include developing tools that can detect and remove bias from the data that are used to train AI algorithms. As a strong advocate of ethical AI, Margaret had developed a tool named Model Cards while at Google, which evaluated the strengths and weaknesses of an AI algorithm. Hugging Face has long been recommending Mitchell’s tool to users. Her joining the team will surely boost the company’s efforts towards removing bias from AI algorithms.
Investors betting on its potential
Seeing the growth that Hugging Face has witnessed since its launch, and the kind of impact NLP will have in the future, the investors are betting on Hugging Face from the beginning. In March this year, Hugging Face raised $40 million in Series B funding led by Addition. In December 2019, the startup had raised $15 million in a Series A funding round led by Lux Capital. It has raised over $60 million till now as per Crunchbase data.
