Home

Awesome

surrogate_module_learning

Code for surrogate module learning (SML) Create a new backward path for more accurate SNN gradients.

Prerequisites

The Following Setup is tested and it is working:

Description

python -m torch.distributed.launch --nproc_per_node=2 --use_env Train_distribute_pallel.py \
   --batch-size 128 --cos_lr_T 300 --epochs 300 \
   --model ResNet_SB18 \
   --num_classes 100 --dataset cifar100 --T 2 \
   --sync-bn --optimizer adamw --lr 0.01 --weight-decay 0.02

Pre-trained models

Citation

Reference paper.