SimCSE在中文任务上的简单实验

Related tags

MiscellaneousSimCSE
Overview

SimCSE 中文测试

SimCSE在常见中文数据集上的测试,包含ATECBQLCQMCPAWSXSTS-B共5个任务。

介绍

文件

- utils.py  工具函数
- eval.py  评测主文件

评测

命令格式:

python eval.py [model_type] [pooling] [task_name] [dropout_rate]

使用例子:

python eval.py BERT cls ATEC 0.3

其中四个参数必须传入,含义分别如下:

- model_type: 模型,必须是['BERT', 'RoBERTa', 'WoBERT', 'RoFormer', 'BERT-large', 'RoBERTa-large', 'SimBERT', 'SimBERT-tiny', 'SimBERT-small']之一;
- pooling: 池化方式,必须是['first-last-avg', 'last-avg', 'cls', 'pooler']之一;
- task_name: 评测数据集,必须是['ATEC', 'BQ', 'LCQMC', 'PAWSX', 'STS-B']之一;
- dropout_rate: 浮点数,dropout的比例,如果为0则不dropout;

环境

测试环境:tensorflow 1.14 + keras 2.3.1 + bert4keras 0.10.5,如果在其他环境组合下报错,请根据错误信息自行调整代码。

下载

Google官方的两个BERT模型:

关于语义相似度数据集,可以从数据集对应的链接自行下载,也可以从作者提供的百度云链接下载。

其中senteval_cn目录是评测数据集汇总,senteval_cn.zip是senteval目录的打包,两者下其一就好。

相关

交流

QQ交流群:808623966,微信群请加机器人微信号spaces_ac_cn

Owner
苏剑林(Jianlin Su)
科学爱好者
苏剑林(Jianlin Su)
Telegram bot for Urban Dictionary.

Urban Dictionary Bot @TheUrbanDictBot A star ⭐ from you means a lot to us! Telegram bot for Urban Dictionary. Usage Deploy to Heroku Tap on above butt

Stark Bots 17 Nov 24, 2022
Run PD patches in NRT using Python

The files in this repository demonstrate how to use Pure Data (Pd) patches designed to run in Non-Real-Time mode to batch-process (synthesize, analyze, etc) sounds in series using Python.

Jose Henrique Padovani 3 Feb 08, 2022
An interactive tool with which to explore the possible imaging performance of candidate ngEHT architectures.

ngEHTexplorer An interactive tool with which to explore the possible imaging performance of candidate ngEHT architectures. Welcome! ngEHTexplorer is a

Avery Broderick 7 Jan 28, 2022
A class to draw curves expressed as L-System production rules

A class to draw curves expressed as L-System production rules

Juna Salviati 6 Sep 09, 2022
FBChecker Account using python , package requests and web old facebook

fbcek FBChecker Account using python , package requests and web old facebook using python 3.x apt upgrade -y apt update -y pkg install bash -y pkg ins

XnuxersXploitXen 5 Dec 24, 2022
Coursework project for DIP class. The goal is to use vision to guide the Dashgo robot through two traffic cones in bright color.

Coursework project for DIP class. The goal is to use vision to guide the Dashgo robot through two traffic cones in bright color.

Yueqian Liu 3 Oct 24, 2022
tgEasy | Easy for a Brighter Shine | Monkey Patcher Addon for Pyrogram

tgEasy | Easy for a Brighter Shine | Monkey Patcher Addon for Pyrogram

Jayant Hegde Kageri 35 Nov 12, 2022
Simple macOS StatusBar app to remind you to unplug your laptop when sufficiently charged

ChargeMon Simple macOS StatusBar app to monitor battery charge status and remind you to unplug your Mac when the battery is sufficiently charged Overv

Rhet Turnbull 5 Jan 25, 2022
An example using debezium and mysql with docker-compose

debezium-mysql An example using debezium and mysql with docker-compose The docker compose starts the Zookeeper, Kafka, Mysql and Debezium Connect. Aft

Horácio Dias Baptista Neto 4 May 21, 2022
Dockernized ZeroTierOne controller with zero-ui web interface.

docker-zerotier-controller Dockernized ZeroTierOne controller with zero-ui web interface. 中文讨论 Customize ZeroTierOne's controller planets Modify patch

sbilly 209 Jan 04, 2023
Python interface to IEX and IEX cloud APIs

Python interface to IEX Cloud Referral Please subscribe to IEX Cloud using this referral code. Getting Started Install Install from pip pip install py

IEX Cloud 41 Dec 21, 2022
Learn to code in any language. If

Learn to Code It is an intiiative undertaken by Student Ambassadors Club, Jamshoro for students who are absolute begineers in programming and want to

Student Ambassadors' Club at Mehran UET 15 Oct 19, 2022
A project to explore and provide useful code for Mango Markets

🥭 Mango Explorer A project to explore and provide useful code for Mango Markets

Blockworks Foundation 160 Dec 19, 2022
Um Script De Mensagem anonimas Para linux e Termux Feito em python

Um Script De Mensagem anonimas Para linux e Termux Feito em python feito em um celular

6 Sep 09, 2021
Providing a working, flexible, easier and faster installer than the one officially provided by Arch Linux

Purpose The purpose is to bring more people to Arch Linux by providing a working, flexible, easier and faster installer than the one officially provid

André Luís 0 Nov 09, 2022
Estimating the potential photovoltaic production of buildings (in Berlin)

The following people contributed equally to this repository (in alphabetical order): Daniel Bumke JJX Corstiaen Versteegh This repository is forked on

Daniel Bumke 6 Feb 18, 2022
A chain of stores wants a 3-month demand forecast for its 10 different stores and 50 different products.

Demand Forecasting Objective A chain store wants a machine learning project for a 3-month demand forecast for 10 different stores and 50 different pro

2 Jan 06, 2022
This is a Blender 2.9 script for importing mixamo Models to Godot-3

Mixamo-To-Godot This is a Blender 2.9 script for importing mixamo Models to Godot-3 The script does the following things Imports the mixamo models fro

8 Sep 02, 2022
Batch obfuscator based on the obfuscation method used by the trick bot launcher

Batch obfuscator based on the obfuscation method used by the trick bot launcher

SlizBinksman 2 Mar 19, 2022
A basic ticketing software.

Ticketer A basic ticketing software. Screenshots Program Launched Issuing Ticket Show your Ticket Entry Done Program Exited Code Features to implement

Samyak Jain 2 Feb 10, 2022