gordicaleksa/pytorch-original-transformer
A PyTorch implementation of the original Transformer architecture from the seminal Vaswani et al. paper, structured as a learning resource.
Not currently ranked — collecting fresh signals.
star history
This repository provides a clean, well-commented PyTorch implementation of the original Transformer model as described in the Attention Is All You Need paper. The code includes educational visualizations in playground.py for concepts like positional encodings and attention mechanisms. It ships with IWSLT pretrained models and is aimed at developers wanting to understand how transformers work under the hood.
Frequently asked
- What is gordicaleksa/pytorch-original-transformer?
- A PyTorch implementation of the original Transformer architecture from the seminal Vaswani et al. paper, structured as a learning resource.
- Is pytorch-original-transformer open source?
- Yes — gordicaleksa/pytorch-original-transformer is open source, released under the MIT license.
- What language is pytorch-original-transformer written in?
- gordicaleksa/pytorch-original-transformer is primarily written in Jupyter Notebook.
- How popular is pytorch-original-transformer?
- gordicaleksa/pytorch-original-transformer has 1.1k stars on GitHub.
- Where can I find pytorch-original-transformer?
- gordicaleksa/pytorch-original-transformer is on GitHub at https://github.com/gordicaleksa/pytorch-original-transformer.