320-rehearsal-free-continual-language-learning-via-efficient-parameter-isolation
Science Score: 18.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
○codemeta.json file
-
○.zenodo.json file
-
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (4.5%) to scientific vocabulary
Last synced: 10 months ago
·
JSON representation
·
Repository
Basic Info
- Host: GitHub
- Owner: SZU-AdvTech-2023
- License: mit
- Language: Python
- Default Branch: main
- Size: 198 KB
Statistics
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
- Releases: 0
Created over 2 years ago
· Last pushed over 2 years ago
Metadata Files
Citation
https://github.com/SZU-AdvTech-2023/320-Rehearsal-Free-Continual-Language-Learning-via-Efficient-Parameter-Isolation/blob/main/
# Environment
Our experimental environment setup is listed below:
* Graphic Card: NVIDIA GeForce RTX 4090
* Processor: Intel(R) Xeon(R) Platinum 8358P CPU @ 2.60GHz
* Operating System: Ubuntu 20.04
* Container: Docker version 24.0.2
* Python Version: 3.7
* Python Libraries: transformers 4.23.1, datasets 2.1.0, pytorch 1.11.0
# Dataset
Download 5ds dataset from [link](https://drive.google.com/file/d/1rWcgnVcNpwxmBI3c5ovNx-E8XKOEL77S/view).
Extract contents from 'LAMOL.tar.gz' to 'data' directory, and run command below.
```sh
python preprocess.py
```
# Run
The syntax of command is listed below:
```sh
python run.py \
--dataset {5ds|WOS} \
--prompt_fusion_mode {None|fusion|last|mean|SMD} \
--method epi \
--query_mode mahalanobis \
--prompt_mode prefix \
--pre_seq_len 32 \
--n_per_class {2000|10|20|50} \
--batch_size 32 \
--lr 0.05 \
--epochs 5
```
Owner
- Name: SZU-AdvTech-2023
- Login: SZU-AdvTech-2023
- Kind: organization
- Repositories: 1
- Profile: https://github.com/SZU-AdvTech-2023
Citation (citation.txt)
@inproceedings{REPO320,
author = "Wang, Zhicheng and Liu, Yufang and Ji, Tao and Wang, Xiaoling and Wu, Yuanbin and Jiang, Congcong and Chao, Ye and Han, Zhencong and Wang, Ling and Shao, Xu and Zeng, Wenqiu",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
pages = "10933--10946",
title = "{Rehearsal-free Continual Language Learning via Efficient Parameter Isolation}",
year = "2023"
}