stable-continual-learning icon indicating copy to clipboard operation
stable-continual-learning copied to clipboard

Change the hyper-parameters in split-CIFAR

Open andrewliao11 opened this issue 3 years ago • 0 comments

In the scripts replicate_experiment_2.sh here, to run the split-CIFAR experiments, the command is

python -m stable_sgd.main --dataset cifar100 --tasks 20 --epochs-per-task 1 --lr 0.1 --gamma 0.8 --hiddens 256 --batch-size 10 --dropout 0.5 --seed 1234

which only provides results around 2x% average accuracy

I found in a closed issue here that the setting should be

python -m stable_sgd.main --dataset cifar100 --tasks 20 --epochs-per-task 1 --lr 0.15 --gamma 0.85 --batch-size 10 --dropout 0.1 --seed 1234

Maybe the scripts need some updates @imirzadeh?

andrewliao11 avatar Feb 13 '21 04:02 andrewliao11