DOP-Tuning(Domain-Oriented Prefix-tuning model)

Overview

DOP-Tuning

DOP-Tuning(Domain-Oriented Prefix-tuning model)代码基于Prefix-Tuning改进.

Files

├── seq2seq                       # Code for encoder-decoder architecture
│   ├── train_bart.py             # high-level scripts to train.
│   ├── prefixTuning.py           # code that implements prefix-tuning.
│   ├── finetune.py               # training code (contains data loading, model loading, and calls trainer)   
│   ├── lightning_base.py         # helper code
│   ├── utils.py                  # helper code
│   └── callbacks.py              # helper code
└── ...

To run the code for encoder-decoder architecture like BART, the code is in seq2seq. This corresponds to the summarization experiments in the paper.

Setup:

cd transformer; pip install -e .

Train via DOP-tuning

cd seq2seq; 

python train_bart.py --mode xsum --preseqlen 200 --do_train yes --fp16 yes --bsz 16  --epoch 30  --gradient_accumulation_step 3 --learning_rate 0.00005  --mid_dim 800 --use_lowdata_token 'yes' --lowdata_token 'summarize'

其中use_lowdata_token表示是否采用domain word初始化的方式;lowdata_token表示传入的domain word.

Decode

cd seq2seq; 

python train_bart.py --mode xsum --do_train no --prefix_model_path {checkpoint_path} --preseqlen {same as training} --mid_dim {same as training} --use_lowdata_token 'yes' --lowdata_token 'summarize'
Owner
Andrew Zeng
Andrew Zeng
Andrew Zeng
An application to see if your Ethereum staking validator(s) are members of the current or next post-Altair sync committees.

eth_sync_committee.py Since the Altair upgrade, 512 validators are randomly chosen every 256 epochs (~27 hours) to form a sync committee. Validators i

4 Oct 27, 2022
Python code for YouTube videos.

#This is a open source project. Python 3 These files are mainly intended to accompany my series of YouTube tutorial videos here, https://www.youtube.c

Joe James 1.3k Dec 26, 2022
My tools box script for sigma

sigma_python_toolbox My tools box script for sigma purpose My goal is not to replace sigma but to put at disposal the scripts that I think to help me

4 Jun 20, 2022
A general illumination correction method for optical microscopy.

CIDRE About CIDRE is a retrospective illumination correction method for optical microscopy. It is designed to correct collections of images by buildin

Kevin Smith 31 Sep 07, 2022
A Tandy Color Computer 1, 2, and 3 assembler written in Python

CoCo Assembler and File Utility Table of Contents What is it? Requirements License Installing Assembler Assembler Usage Input File Format Print Symbol

Craig Thomas 16 Nov 03, 2022
A QGIS integration plugin for Kart repositories

QGIS Kart Plugin A plugin to work with Kart repositories Installation The Kart plugin is available in the QGIS Plugins server. To install the latest v

Koordinates 27 Jan 04, 2023
Synthetik Python Mod - A save editor tool for the game Synthetik written in python

Synthetik_Python_Mod A save editor tool for the game Synthetik written in python

2 Sep 10, 2022
Pdraw - Generate Deterministic, Procedural Artwork from Arbitrary Text

pdraw.py: Generate Deterministic, Procedural Artwork from Arbitrary Text pdraw a

Brian Schrader 2 Sep 12, 2022
SQL centered, docker process running game

REQUIREMENTS Linux Docker Python/bash set up image "docker build -t game ." create db container "run my_whatever/game_docker/pdb create" # creating po

1 Jan 11, 2022
This is a Docker-based pipeline for preparing sextractor-ready multiwavelength images

Pipeline for creating NB422-detected (ODI) catalog The repository contains a Docker-based pipeline for preprocessing observational data. The pipeline

1 Sep 01, 2022
An early stage integration of Hotwire Turbo with Django

Note: This is not ready for production. APIs likely to change dramatically. Please drop by our Slack channel to discuss!

Hotwire for Django 352 Jan 06, 2023
Python project that aims to discover CDP neighbors and map their Layer-2 topology within a shareable medium like Visio or Draw.io.

Python project that aims to discover CDP neighbors and map their Layer-2 topology within a shareable medium like Visio or Draw.io.

3 Feb 11, 2022
Простенький ботик для троллинга с интерфейсом #Yakima_Visus

Bot-Trolling-Vk Простенький ботик для троллинга с интерфейсом #Yakima_Visus Установка pip install vk_api pip install requests если там еще чото будет

Yakima Visus 4 Oct 11, 2022
TeamFleming is a multicultural group of 20 young bioinformatics enthusiasts participating in the 2021 HackBio Virtual Summer Internship

💻 Welcome to Team Fleming's Repo! #TeamFleming is a multicultural group of 20 young bioinformatics enthusiasts participating in the 2021 HackBio Virt

3 Aug 08, 2021
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
Write Streamlit apps using Notion! (Prototype)

Streamlit + Notion test app Write Streamlit apps using Notion! ☠️ IMPORTANT: This is just a little prototype I made to play with some ideas. Not meant

Thiago Teixeira 22 Sep 08, 2022
Uma versão em Python/Ursina do aplicativo Real Drum (android).

Real Drum Descrição Esta é uma versão alternativa feita em Python com a engine Ursina do aplicatio Real Drum (presente no Google Play Store). Como exe

hayukimori 5 Aug 20, 2022
Automatically deletes Capital One Eno virtual cards for when you've made a couple too many.

eno-delete Automatically deletes Capital One Eno virtual cards for when you've made a couple too many. Warning: Program will delete ALL virtual cards

h3x 3 Sep 29, 2022
Verification of Monty Hall problem by experimental simulation.

Verification of Monty Hall problem by experimental simulation. |中文|English| In the process of learning causal inference, I learned about the Monty Hal

云端听茗 1 Nov 22, 2022
Basic infrastructure for writing scripts in Python

Base Script Python is an excellent language that makes writing scripts very straightforward. Over the course of writing many scripts, we realized that

Deep Compute, LLC 9 Jan 07, 2023