p-tuning_NLU

Overview

这个小项目是受乐于分享的苏剑林大佬这篇p-tuning 文章启发，也实现了个使用P-tuning进行NLU分类的任务，思路是一样的，prompt实现方式有不同，这里是将[unused*]的embeddings参数抽取出用于初始化prompt_embed后，再接一个lstm和mlp用于关联各prompt，与最初p-tuning提出《GPT Understands, Too》的实现一样，结果显示在few-shot上p-tuning非常接近finetune效果。

Dataset

数据是情感分类，下载地址百度网盘提取码：osja

Evaluation

1. finetune

python few_shot_finetune.py

测试集效果：

epoch: 0 - acc: 0.897679 - best_test_acc: 0.8976788252013264
epoch: 1 - acc: 0.876362 - best_test_acc: 0.8976788252013264
epoch: 2 - acc: 0.884889 - best_test_acc: 0.8976788252013264
epoch: 3 - acc: 0.884415 - best_test_acc: 0.8976788252013264
epoch: 4 - acc: 0.884415 - best_test_acc: 0.8976788252013264

全量参数对小样本进行finetune，仅1个epoch就收敛了

2. p-tuning

python few_shot_ptuning.py

测试集效果：

epoch: 0 - acc: 0.546660 - best_test_acc: 0.5466603505447655
epoch: 1 - acc: 0.687826 - best_test_acc: 0.6878256750355282
epoch: 2 - acc: 0.737091 - best_test_acc: 0.7370914258645191
epoch: 3 - acc: 0.722406 - best_test_acc: 0.7370914258645191
epoch: 4 - acc: 0.776883 - best_test_acc: 0.7768829938417812
epoch: 5 - acc: 0.805306 - best_test_acc: 0.8053055423969683
epoch: 6 - acc: 0.833254 - best_test_acc: 0.8332543818095689
epoch: 7 - acc: 0.837991 - best_test_acc: 0.8379914732354334
epoch: 8 - acc: 0.854571 - best_test_acc: 0.8545712932259593
epoch: 9 - acc: 0.858361 - best_test_acc: 0.8583609663666508
epoch: 10 - acc: 0.856466 - best_test_acc: 0.8583609663666508
epoch: 11 - acc: 0.853150 - best_test_acc: 0.8583609663666508
epoch: 12 - acc: 0.868783 - best_test_acc: 0.8687825675035529
epoch: 13 - acc: 0.877309 - best_test_acc: 0.877309332070109
epoch: 14 - acc: 0.873993 - best_test_acc: 0.877309332070109
epoch: 15 - acc: 0.877783 - best_test_acc: 0.8777830412126955
epoch: 16 - acc: 0.882994 - best_test_acc: 0.8829938417811464
epoch: 17 - acc: 0.881573 - best_test_acc: 0.8829938417811464
epoch: 18 - acc: 0.889626 - best_test_acc: 0.8896257697773567
epoch: 19 - acc: 0.877783 - best_test_acc: 0.8896257697773567

仅prompt_embed和lstm及mlp去做p-tuning，20个epoch后接近收敛，acc=0.8896，略小于finetun的acc 0.8977

附上苏神结果对比：

p-tuning for few-shot NLU task

Related tags

Overview

p-tuning_NLU

Overview

Dataset

Evaluation

Owner

Predicting the usefulness of reviews given the review text and metadata surrounding the reviews.

This is my reading list for my PhD in AI, NLP, Deep Learning and more.

Guide to using pre-trained large language models of source code

CorNet Correlation Networks for Extreme Multi-label Text Classification

Chinese Grammatical Error Diagnosis

Sentence Embeddings with BERT & XLNet

GNES enables large-scale index and semantic search for text-to-text, image-to-image, video-to-video and any-to-any content form

Source code of the "Graph-Bert: Only Attention is Needed for Learning Graph Representations" paper

Persian-lexicon - A lexicon of 70K unique Persian (Farsi) words

justCTF [*] 2020 challenges sources

Basic yet complete Machine Learning pipeline for NLP tasks

Simple python code to fix your combo list by removing any text after a separator or removing duplicate combos

Chinese named entity recognization (bert/roberta/macbert/bert_wwm with Keras)

基于GRU网络的句子判断程序/A program based on GRU network for judging sentences

Contains links to publicly available datasets for modeling health outcomes using speech and language.

Using context-free grammar formalism to parse English sentences to determine their structure to help computer to better understand the meaning of the sentence.

Twitter-Sentiment-Analysis - Twitter sentiment analysis for india's top online retailers(2019 to 2022)

Simple text to phones converter for multiple languages

Rethinking the Truly Unsupervised Image-to-Image Translation - Official PyTorch Implementation (ICCV 2021)

Code for "Generative adversarial networks for reconstructing natural images from brain activity".