leeyy2020 Goto Github PK

adapet

[EMNLP 2021] Improving and Simplifying Pattern Exploiting Training

leeyy.github.io

lm-bff

ACL'2021: LM-BFF: Better Few-shot Fine-tuning of Language Models

logic

nccl

Sample examples of how to call collective operation functions on multi-GPU environments. A simple example of using broadcast, reduce, allGather, reduceScatter and sendRecv operations.

p-tuning

A novel method to tune language models. Codes and datasets for paper ``GPT understands, too''.

paddle

PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice （『飞桨』核心框架，深度学习&机器学习高性能单机、分布式训练和跨平台部署）

paddlenlp

An NLP library with Awesome pre-trained Transformer models and easy-to-use interface, supporting wide-range of NLP tasks from research to industrial applications.

parsha256_gpu

parsha256_gpu-1

pet

pet-1

This repository contains the code for "Exploiting Cloze Questions for Few-Shot Text Classification and Natural Language Inference"

pytorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration

qe

upload

warp

Code for ACL'2021 paper WARP 🌀 Word-level Adversarial ReProgramming. Outperforming `GPT-3` on SuperGLUE Few-Shot text classification. https://aclanthology.org/2021.acl-long.381/