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)
Lightweight Scheduled Blocks Checker for Current Epoch. No cardano-node Required, data is taken from blockfrost.io

ReLeaderLogs For Cardano Stakepool Operators: Lightweight Scheduled Blocks Checker for Current Epoch. No cardano-node Required, data is taken from blo

SNAKE (Cardano Stakepool) 2 Oct 19, 2021
Framework for creating efficient data processing pipelines

Aqueduct Framework for creating efficient data processing pipelines. Contact Feel free to ask questions in telegram t.me/avito-ml Key Features Increas

avito.tech 137 Dec 29, 2022
IPython: Productive Interactive Computing

IPython: Productive Interactive Computing Overview Welcome to IPython. Our full documentation is available on ipython.readthedocs.io and contains info

IPython 15.6k Dec 31, 2022
Python solution of advent-of-code 2021

Advent of code 2021 Python solutions of Advent of Code 2021 written by Eric Bouteillon Requirements The solutions were developed and tested using Pyth

Eric Bouteillon 3 Oct 25, 2022
Files for QMC Workshop 2021

QMC Workshop 2021 This repository contains the presented slides and example files for the Quantum Monte Carlo (QMC) Workshop 5 October - 23 November,

QMCPACK 39 Nov 04, 2022
Interactive class notebooks for ECE4076 Computer Vision, weeks 1 - 6

ECE4076 Interactive class notebooks for ECE4076 Computer Vision, weeks 1 - 6. ECE4076 is a computer vision unit at Monash University, covering both cl

Michael Burke 9 Jun 16, 2022
Pre-1.0 door/chest sound injector for Minecraft

doorjector Pre-1.0 door/chest sound injector for Minecraft. While the game is running, doorjector hotswaps the new sounds for the old right before the

Sam 1 Nov 20, 2021
Have an idea for a Python package? Register the name on PyPI 💡

Register Package Names on PyPI Have an idea for a Python package? Thought of a great name? Register it on PyPI, before someone else does! A tool that

Alex Ioannides 1 Jul 15, 2022
An application for automation of the mining function in the game Alienworlds.IO

alienautomation A Python script made to automate the tidious job of mining on AlienWorlds This script: Automatically opens the browser Automatically l

anonieXdev 42 Dec 03, 2022
A simple streamlit webapp with multiple functionality

A simple streamlit webapp with multiple functionality

Omkar Pramod Hankare 2 Nov 24, 2021
Service for working with open data of the State Duma of the Russian Federation

Сервис для работы с открытыми данными Госдумы РФ Исходные данные из API Госдумы РФ извлекаются с помощью Apache Nifi и приземляются в хранилище Clickh

Aleksandr Sergeenko 2 Feb 14, 2022
A simple wrapper to analyse and visualise reinforcement learning agents' behaviour in the environment.

Visrl Visrl (pronounced "visceral") is a simple wrapper to analyse and visualise reinforcement learning agents' behaviour in the environment. Reinforc

Jet New 14 Jun 27, 2022
This is a simple leaderboard for 30 days of Google Cloud program for students of ASIET

30daysleaderboard #Hacktoberfest - Please don't make changes in readme file. Only improvement in the project will be accepted. Update - Now if you run

5 Oct 29, 2021
An implementation of Ray Tracing in One Weekend using Taichi

又一个Taichi语言的Ray Tracer 背景简介 这个Ray Tracer基本上是照搬了Peter Shirley的第一本小书Ray Tracing in One Weekend,在我写的时候参考的是Version 3.2.3这个版本。应该比其他中文博客删改了不少内容。果然Peter Shir

张皓 30 Nov 21, 2022
A library for pattern matching on symbolic expressions in Python.

MatchPy is a library for pattern matching on symbolic expressions in Python. Work in progress Installation MatchPy is available via PyPI, and

High-Performance and Automatic Computing 151 Dec 24, 2022
Script to check if your Bistromatic handle everything as it should.

Bistromatic Checker Script to check if your Bistromatic handle everything as it should. The bistromatic is the project marking the end of the CPool at

Mathias 1 Dec 27, 2021
Zeus - Advanced Punishments with Embeds.

Zeus Advanced Punishments with Embeds. Make sure to put the Discord Bot Token in the " TOKEN = '' " Language Python Features Ban Kick Mute Unmute Warn

2 Jan 05, 2022
Anki for desktop computers

Anki This repo contains the source code for the computer version of Anki. If you'd like to try development builds of Anki but don't feel comfortable b

Ankitects 12.9k Jan 09, 2023
Personal Finance Forecaster - An AI tool for forecasting personal expenses

Personal Finance Forecaster - An AI tool for forecasting personal expenses

2 Mar 09, 2022
Cross-platform MachO/ObjC Static binary analysis tool & library. class-dump + otool + lipo + more

ktool Static Mach-O binary metadata analysis tool / information dumper pip3 install k2l Development is currently taking place on the @python3.10 branc

Kritanta 301 Dec 28, 2022