- Notifications
You must be signed in to change notification settings - Fork1
The official PyTorch code for AAAI'23 Paper "Sparse Coding in a Dual Memory System for Lifelong Learning"
License
NotificationsYou must be signed in to change notification settings
NeurAI-Lab/SCoMMER
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
Official Repository for the(Oral) AAAI'23 paperSparse Coding in a Dual Memory System for Lifelong Learning
We extended theCLS-ER repo with our method
Use
python main.pyto run experiments.Use argument
--load_best_argsto use the best hyperparameters for each of the evaluation setting from the paper.To reproduce the results in the paper run the following
python main.py --dataset <dataset> --model <model> --experiment_id <experiment_id> --buffer_size <buffer_size> --load_best_argspython main.py --dataset seq-cifar10 --model scommer --buffer_size 200 --experiment_id scommer-c10-200 --load_best_argspython main.py --dataset seq-cifar100 --model scommer --buffer_size 200 --experiment_id scommer-c100-200 --load_best_argspython main.py --dataset gcil-cifar100 --weight_dist unif --model scommer --buffer_size 200 --experiment_id scommer-gcil-unif-200 --load_best_argspython main.py --dataset gcil-cifar100 --weight_dist longtail --model scommer --buffer_size 200 --experiment_id scommer-gcil-longtail-200 --load_best_args
torch==1.7.0
torchvision==0.9.0
quadprog==0.1.7