Early-Exit with Class Exclusion for Efficient Inference of Neural Networks

Train Model

python --cf train.py ./configs/{alexnet, resnet, vgg}/train/{model_name}_{dataset_name}_BC.yaml

BC for our method, CR for the traditional method.

$\beta$ Search

python fine_tune.py --cf ./configs/{alexnet, resnet, vgg}/search/{model_name}_{dataset_name}_BC.yaml

Test

python exit_text.py --cf ./configs/{alexnet, resnet, vgg}/test/{model_name}_{dataset_name}_BC.yaml

Reference

@INPROCEEDINGS{Wang2023EarlyExitWC,

title={Early-Exit with Class Exclusion for Efficient Inference of Neural Networks},

author={Jingcun Wang and Bing Li and Grace Li Zhang},

booktitle={2024 IEEE 6th International Conference on AI Circuits and Systems (AICAS)},

year={2024} }