From 757bf0dabb9b95b3ade106ddee0dbd57cee5a8e0 Mon Sep 17 00:00:00 2001 From: matteo-grella Date: Mon, 30 Oct 2023 22:58:59 +0100 Subject: [PATCH] Update README.md --- README.md | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/README.md b/README.md index bd59135..5a7c759 100644 --- a/README.md +++ b/README.md @@ -1,14 +1,10 @@ # Cybertron -Cybertron is a pure Go package that provides a simple and easy-to-use interface for cutting-edge Natural Language Processing (NLP) technologies. +Cybertron is a Go package designed for developers to easily implement NLP using Transformer models, such as BERT and BART. +It's tailored for inference tasks and allows for simple server deployment with a single executable. +This tool leverages pre-trained models from the [HuggingFace models repository](https://huggingface.co/models), converting models for use with the Go-based [Spago](https://github.com/nlpodyssey/spago) framework. While spaGO is a complete ML framework that may eventually allow for model fine-tuning, Cybertron's current focus is on **inference**. -It enables Go developers to use state-of-the-art neural technologies i.e. Transformers, without having to learn other languages or worry about heavy deep learning frameworks (thus, the deployment is just a single executable for your server!). - -Luckily, pre-trained /fine-tuned Transformer models exist for several languages and are publicly hosted on the [HuggingFace models repository](https://huggingface.co/models). - -A unique feature of Cybertron is its compatibility with [HuggingFace Transformers](https://github.com/huggingface/transformers): it can run **inference** on PyTorch pre-trained models after they have been automatically downloaded and converted to the [Spago](https://github.com/nlpodyssey/spago) format. - -> Cybertron currently supports a few architectures (BERT, BART, and derivatives), and we're seeking collaborators to speed up its development! +The project is open for collaboration to expand its development further. ## Supported tasks