• Implementation for the paper "Efficient Channel Pruning Based on Probability Model Bypassing and Architecture Alignment"
• • Overview
  • HOWTO
  • • Search on Different Architecture
    • Evaluate the Model Configuration Found by Search
  • Bibliography
  • Acknowledgement

Implementation for the paper "Efficient Channel Pruning Based on Probability Model Bypassing and Architecture Alignment"

Overview

├── checkpoints (Pretrained super-net checkpoints)
│   └── stage1
│       ├── mbv2-stage1.pth
│       ├── res18-stage1.pth
│       └── res50-stage1.pth
├── config (Configuration snippets)
│   └── imagenet
│       ├── mbv2
│       │   └── 8gpu
│       │       └── 2xfloss.yaml
│       └── res50
│           └── 8gpu
│               └── 2xfloss.yaml
├── eval (Evaluation root path)
│   ├── models
│   │   ├── adaptive_mobilenet.py
│   │   ├── adaptive_resnet.py
│   │   ├── amc_mbv2.py
│   │   ├── inceptionresnetv2.py
│   ├── retrain_no_sync.py
│   └── retrain_res50.py
├── models
│   ├── adaptive (Configurable model for evaluation)
│   │   ├── __init__.py
│   │   ├── mobilenet.py
│   │   └── resnet.py
│   ├── ecp (Super-net model for ECP) 
│   │   ├── alpha_op.py
│   │   ├── ecp_mobilenet.py
│   │   ├── ecp_resnet.py
│   │   ├── __init__.py
│   │   └── utils.py
│   ├── __init__.py
│   └── slimmable
│       ├── __init__.py
│       ├── us_mobilenet.py
│       ├── us_ops.py
│       └── us_resnet.py
├── runner (APIs for executing training and search for different methods)
│   ├── ecp_runner.py
│   ├── __init__.py
│   ├── normal_runner.py
│   └── us_runner.py
└── utils (Utility functions)
    ├── data.py
    ├── distributed.py
    ├── __init__.py
    ├── loss_fn.py
    ├── lr_scheduler.py
    ├── meter.py
    └── tools.py
├── search-example.sh (Execution script for search)
├── eval-example.sh  (Execution script for evaluation) 
├── main.py (Universal entry point)
├── requirements.txt (dependencies)

HOWTO

Search on Different Architecture

bash search-example.sh

Evaluate the Model Configuration Found by Search

bash eval-example.sh

Bibliography

If you find the code in this repository useful in your research, please consider citing our paper.

@inproceedings{ECP,
	address = {USA},
	title = {Efficient {Channel} {Pruning} {Based} on {Architecture} {Alignment} and {Probability} {Model} {Bypassing}},
	doi = {10.1109/SMC52423.2021.9659289},
	language = {en},
	booktitle = {Proceedings of the {IEEE} {International} {Conference} on {Systems}, {Man}, and {Cybernetics}},
	publisher = {IEEE},
	author = {Tao, Lvfang and Gao, Wei},
	year = {2021},
	pages = {3232--3237},
}

Acknowledgement

This implementation was based on the following works:

• DMCP (https://github.com/Zx55/dmcp)