/AGVM

Large-batch Optimization for Dense Visual Predictions (NeurIPS 2022)

Primary LanguagePythonApache License 2.0Apache-2.0

AGVM

This repo is the official implementation of "Large-batch Optimization for Dense Visual Predictions (NeurIPS 2022)". Since we adopted private frameworks (POD and LinkLink) to conduct the experiments previously, the results open-sourced with mmdetection will be slightly different from the results in our paper. The optimized version of DDP will be released in the future.

Usage

Installation

Step 0. Please refer to mmdetection get started for installation and dataset preparation.

Step 1. Install AGVM from source:

git clone https://github.com/Sense-X/AGVM.git
cd AGVM
make install

Training

Please refer to this doc for examples of training.

Results

The box mAP of Faster R-CNN:

Batch Size 32 256 512
Baseline 37.1 36.7 36.2
AGVM - 37.1 (config) 36.8 (config)

The seg mAP of Mask R-CNN:

Batch Size 32 256 512
Baseline 34.8 34.4 33.9
AGVM - 35.0 (config) 34.6 (config)

Citation

@article{xue2022large,
  title = {Large-batch Optimization for Dense Visual Predictions},
  author = {Zeyue Xue and Jianming Liang and Guanglu Song and Zhuofan Zong and Liang Chen and Yu Liu and Ping Luo},
  year = {2022},
  journal = {arXiv:2210.11078}
}