Skip to content

INFOCOM 2024: Online Resource Allocation for Edge Intelligence with Colocated Model Retraining and Inference

License

Notifications You must be signed in to change notification settings

caihuaiguang/ORRIC

Repository files navigation

Online Resource Allocation for Edge Intelligence with Colocated Model Retraining and Inference

Prerequisites

  • pytorch, numpy, thop

Usage

This is the implementation of ORRIC using the trained model in folder models.

The corresponding dataset can be found at https://zenodo.org/records/2535967. Please decompress CIFAR-10-C.tar in the root directory to obtain the CIFAR-10-C folder.

# ORRIC implementation
python train_inference_two_model.py 

Others:

# teacher (resnet_50) training
python res50_teacher_training.py
# student (mobilenet_v2) training
python mobile_student_training.py

# measure the MACs
python MACs.py
# measure the time
python measure_time.py
# measure the performance
python performance.py

Slides

Slides

Citation

@INPROCEEDINGS{cai2024ORRIC,
  author={Huaiguang Cai and
          Zhi Zhou and
          Qianyi Huang},
  booktitle={IEEE INFOCOM 2024 - IEEE Conference on Computer Communications}, 
  title={Online Resource Allocation for Edge Intelligence with Colocated Model Retraining and Inference}, 
  year={2024},
  pages={1900-1909},
  doi={10.1109/INFOCOM52122.2024.10621206}}

License

This project is licensed under the MIT License - see the LICENSE file for details

About

INFOCOM 2024: Online Resource Allocation for Edge Intelligence with Colocated Model Retraining and Inference

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages