CCRP

Description: This package includes the Python code of the CCRP algorithm [1] for robust neural network pruning. In CCRP, the robust neural network pruning is formulated as a three-objective optimization problem and solved by a cooperative coevolution pruning framework. Experiments on two datasets CIFAR-10 and SVHN show the good performance of CCRP. README files are included in the package, showing how to use the code.

References: [1] Jia-Liang Wu, Haopu Shang, Wenjing Hong, and Chao Qian. Robust Neural Network Pruning by Cooperative Coevolution. In: Proceedings of the 17th International Conference on Parallel Problem Solving from Nature (PPSN'22), Dortmund, Germany, 2022.

ATTN: This package is free for academic usage. You can run it at your own risk. For other purposes, please contact Dr. Chao Qian (qianc@lamda.nju.edu.cn).

Requirement: The package was developed with Python, especially on PyTorch.

ATTN2: This package was developed by Mr. Jia-Liang Wu (wujl@lamda.nju.edu.cn). For any problem concerning the code, please feel free to contact Mr. Wu.

Download: code (48MB)