YaFSDP: Yet another Fully Sharded Data Parallel

Cramming the training of a (BERT-type) language model into limited compute.

🗃️ A Python ORM-like interface for the Clingo Answer Set Programming (ASP) reasoner

An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.

Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.

Fine-tuning LLMs using QLoRA

OpenChat: Advancing Open-source Language Models with Imperfect Data

A fast inference library for running LLMs locally on modern consumer-class GPUs

distributed trainer for LLMs

Efficient Attention for Long Sequence Processing

A high-throughput and memory-efficient inference and serving engine for LLMs

Fast and memory-efficient exact attention

Kernl lets you run PyTorch transformer models several times faster on GPU with a single line of code, and is designed to be easily hackable.

QLoRA: Efficient Finetuning of Quantized LLMs

A part of options trading educational platform project implemented for HackNYU 2023.

The most lightweight python docker image possible

Resources, datasets, papers on Question Answering

Tools for checking ACL paper submissions

MT Evaluation in Many Languages via Zero-Shot Paraphrasing

Facebook Low Resource (FLoRes) MT Benchmark

Multi-container environment with Hadoop, Spark and Hive

Theano code for experiments in the paper "A Hybrid Convolutional Variational Autoencoder for Text Generation."

How to encode sentences in a high-dimensional vector space, a.k.a., sentence embedding.

ELSA combines extractive and abstractive approaches to the automatic text summarization

A PyTorch implementation of Transformer in "Attention is All You Need"

Tools for extracting tables and results from Machine Learning papers

Quantile-based approach to estimating cognitive text complexity