Official Implementation of EAGLE-1 and EAGLE-2

BERT score for text generation

Unsupervised Word Segmentation for Neural Machine Translation and Text Generation

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

An open-source NLP research library, built on PyTorch.

KenLM: Faster and Smaller Language Model Queries

Google Research

Implementation of the LLaMA language model based on nanoGPT. Supports flash attention, Int8 and GPTQ 4bit quantization, LoRA and LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.

OpenICL is an open-source framework to facilitate research, development, and prototyping of in-context learning.

Reference BLEU implementation that auto-downloads test sets and reports a version string to facilitate cross-lab comparisons

SpikingJelly is an open-source deep learning framework for Spiking Neural Network (SNN) based on PyTorch.

Implementation of Imputer: Sequence Modelling via Imputation and Dynamic Programming in PyTorch

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

A learning rate range test implementation in PyTorch

Deep learning framework to train, deploy, and ship AI products Lightning fast.

Ongoing research training transformer models at scale

The fastai deep learning library

LightSeq: A High Performance Library for Sequence Processing and Generation