SNN_Attention_VGG

Run CIFAR10 with VGG7

python3 CIFAR10_VGG.py --T 1 --model spiking_vgg7 --data-path *Your CIFAR10 Data Path* --batch-size 64 --lr 0.001 --lr-scheduler cosa --epochs 200 --lr-warmup-epochs 0 --opt adamw --train-crop-size 32 --val-crop-size 32 --val-resize-size 32

Run CIFAR10 with VGG7

python3 CIFAR10_VGG_Att.py --T 1 --model spiking_vgg7 --data-path *Your CIFAR10 Data Path* --batch-size 64 --lr 0.001 --lr-scheduler cosa --epochs 200 --lr-warmup-epochs 0 --opt adamw --train-crop-size 32 --val-crop-size 32 --val-resize-size 32

Run CIFAR10 with VGG11

python3 CIFAR10_VGG.py --T 1 --model spiking_vgg11_bn --data-path *Your CIFAR10 Data Path* --batch-size 64 --lr 0.001 --lr-scheduler cosa --epochs 200 --lr-warmup-epochs 0 --opt adamw --train-crop-size 32 --val-crop-size 32 --val-resize-size 32

Run CIFAR10 with VGG11 Attention

python3 CIFAR10_VGG_Att.py --T 1 --model spiking_vgg11_bn --data-path *Your CIFAR10 Data Path* --batch-size 64 --lr 0.001 --lr-scheduler cosa --epochs 200 --lr-warmup-epochs 0 --opt adamw --train-crop-size 32 --val-crop-size 32 --val-resize-size 32

Run CIFAR100 with VGG11

python3 CIFAR100_VGG.py --T 1 --model spiking_vgg11_bn --data-path *Your CIFAR100 Data Path* --batch-size 64 --lr 0.001 --lr-scheduler cosa --epochs 200 --lr-warmup-epochs 0 --opt adamw --train-crop-size 32 --val-crop-size 32 --val-resize-size 32

Run CIFAR100 with VGG11 Attention

python3 CIFAR100_VGG_Att.py --T 1 --model spiking_vgg11_bn --data-path *Your CIFAR100 Data Path* --batch-size 64 --lr 0.001 --lr-scheduler cosa --epochs 200 --lr-warmup-epochs 0 --opt adamw --train-crop-size 32 --val-crop-size 32 --val-resize-size 32

Run CIFAR100 with VGG13

python3 CIFAR100_VGG.py --T 1 --model spiking_vgg13_bn --data-path *Your CIFAR100 Data Path* --batch-size 64 --lr 0.001 --lr-scheduler cosa --epochs 200 --lr-warmup-epochs 0 --opt adamw --train-crop-size 32 --val-crop-size 32 --val-resize-size 32

Run CIFAR100 with VGG13 Attention

python3 CIFAR100_VGG_Att.py --T 1 --model spiking_vgg13_bn --data-path *Your CIFAR100 Data Path* --batch-size 64 --lr 0.001 --lr-scheduler cosa --epochs 200 --lr-warmup-epochs 0 --opt adamw --train-crop-size 32 --val-crop-size 32 --val-resize-size 32

Citation

@ARTICLE{10032591,
  author={Yao, Man and Zhao, Guangshe and Zhang, Hengyu and Hu, Yifan and Deng, Lei and Tian, Yonghong and Xu, Bo and Li, Guoqi},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, 
  title={Attention Spiking Neural Networks}, 
  year={2023},
  volume={},
  number={},
  pages={1-18},
  doi={10.1109/TPAMI.2023.3241201}}