README.md

Recasting Continual Learning as Sequence Modeling

This repository contains the code for our NeurIPS 2023 paper titled Recasting Continual Learning as Sequence Modeling. For a brief overview of the paper, please check this tweet.

Requirements

• Python 3.10
• Pip packages:

pip install -r requirements.txt

Usage

The basic usage of the training script is as follows:

python train.py -mc [model config] -dc [data config] -o [override options] -l [log directory]

In commands.sh, we provide the commands used to train the models in the paper.

Downloading Datasets

All datasets except MS-Celeb-1M are downloaded automatically by the code.

MS-Celeb-1M

Use BitTorrent to download the dataset from Academic Torrents.

transmission-cli https://academictorrents.com/download/9e67eb7cc23c9417f39778a8e06cca5e26196a97.torrent -w data

Citation

@inproceedings{Lee2023Recasting,
  author    = {Soochan Lee and Jaehyeon Son and Gunhee Kim},
  title     = {Recasting Continual Learning as Sequence Modeling},
  booktitle = {NeurIPS},
  year      = {2023},
}