minimal-nmt

Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.

Unsupervised text tokenizer for Neural Network-based text generation.

Open Source Neural Machine Translation and (Large) Language Models in PyTorch

Implementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch

A concise but complete full-attention transformer with a set of promising experimental features from various papers

Graph Attention Networks (https://arxiv.org/abs/1710.10903)

LightSeq: A High Performance Library for Sequence Processing and Generation

Pytorch implementation of the Graph Attention Network model by Veličković et. al (2017, https://arxiv.org/abs/1710.10903)

A PyTorch implementation of Location-Relative Attention Mechanisms For Robust Long-Form Speech Synthesis