Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
PyTorch Implementation of Compact Generalized Non-local Network [NIPS'18]
| Date | Stars |
|---|---|
| 2026-07-30 | 257 |
| 2026-11-04 | 257 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Compact Generalized Non-local Network
By [Kaiyu Yue](http://kaiyuyue.com), [Ming Sun](https://msunming.github.io/), [Yuchen Yuan](https://scholar.google.com.au/citations?user=QJAr1KoAAAAJ&hl=en), [Feng Zhou](http://www.f-zhou.com/bio.html), Errui Ding and Fuxin Xu
## Introduction
This is a [PyTorch](https://pytorch.org/) re-implementation for the paper [Compact Generalized Non-local Network](https://arxiv.org/abs/1810.13125). It brings the CGNL models trained on the [CUB-200](http://www.vision.caltech.edu/visipedia/CUB-200.html), [ImageNet](http://image-net.org/index) and [COCO](http://cocodataset.org/) based on [maskrcnn-benchmark](https://github.com/facebookresearch/maskrcnn-benchmark) from FAIR.
## Update
- 05/2019: Support CGNL & NL Module in Caffe - see [`caffe/README.md`](./caffe/README.md)
## Citation
If you think this code is useful in your research or wish to refer to the baseline results published in our paper, please use the following BibTeX entry.
```
@article{CGNLNetwork2018,
author={Kaiyu Yue and Ming Sun and Yuchen Yuan and Feng Zhou and Errui Ding and Fuxin Xu},
title={Compact Generalized Non-local Network},
journal={NIPS},
year={2018}
}
```
## Requirements
* PyTorch >= 0.4.1 or 1.0 from a nightly release
* Python >= 3.5
* torchvision >= 0.2.1
* termcolor >= 1.1.0
## Environment
The code is developed and tested under 8 Tesla P40 / V100-SXM2-16GB GPUS cards on CentOS with installed CUDA-9.2/8.0 and cuDNN-7.1.
## Baselines and Main Results on CUB-200 Dataset
| File ID | Model | Best Top-1 (%) | Top-5 (%) | Google Drive | Baidu Pan |
|:---------- |:-------------------- |:--------------:|:---------:|:------------:|:---------:|
| 1832260500 | R-50 Base | 86.45 | 97.00 | [`link`](https://drive.google.com/file/d/1nk1y6YE5A-jcuxC5r-hdnU0B5dny8Nf2/view?usp=sharing) | [`link`](https://pan.baidu.com/s/1QB4GLaq0rY9C7psItj8IdA) |
| 1832260501 | R-50 w/ 1 NL Block | 86.69 | 96.95 | [`link`](https://drive.google.com/file/d/1r15lAheWPHyo4V9aitceyG6h-rEPwzun/view?usp=sharing) | [`link`](https://pan.baidu.com/s/1UERXF48oM8MKRfhRFdrHFw) |
| 1832260502 | R-50 w/ 1 CGNL Block | 87.06 | 96.91 | [`link`](https://drive.google.com/file/d/1XND1fPWCzlYTsfiggiZqYhFREAZ-DVTK/view?usp=sharing) | [`link`](https://pan.baidu.com/s/1onJlxgEhQxTVwUdHVmNXRA) |
| 1832261010 | R-101 Base | 86.76 | 96.91 | [`link`](https://drive.google.com/file/d/1mZPz9E3eEwmusEBD-zpjrL6IceM8kxOq/view?usp=sharing) | [`link`](https://pan.baidu.com/s/1NWTbRny-Qs6_7Ts40RkeAQ) |
| 1832261011 | R-101 w/ 1 NL Block | 87.04 | 97.01 | [`link`](https://drive.google.com/file/d/1eYKqnFxSVfCM_5qeTeiEG69PteaQZ3DT/view?usp=sharing) | [`link`](https://pan.baidu.com/s/1JFu5JRjNo9drQZkDeZzPxw) |
| 1832261012 | R-101 w/ 1 CGNL Block | 87.28 | 97.20 | [`link`](https://drive.google.com/file/d/1oiRzne2nhS4aUwLWr3DzS7ke3D6HYMZ2/view?usp=sharing) | [`link`](https://pan.baidu.com/s/1iTMZPd2BuXcyZl56sEWUrA) |
### Notes:
- The input size is 448.
- The CGNL block with dot production kernel is configured within 8 groups.
| File ID | Model | Best Top-1 (%) | Top-5 (%) | Google Drive | Baidu Pan |
|:----------- |:---------------------- |:--------------:|:---------:|:------------:|:---------:|
| 1832260503x | R-50 w/ 1 CGNLx Block | 86.56 | 96.63 | [`link`](https://drive.google.com/file/d/1grkSRJfojIECZs85abvfelnxnVleSLlJ/view?usp=sharing) | [`link`](https://pan.baidu.com/s/1RisiSxWMBjVDnhxeM2Ddaw) |
| 1832261013x | R-101 w/ 1 CGNLx Block | 87.18 | 97.03 | [`link`](https://drive.google.com/file/d/1AZPZk4IvmsbT6jcm0An_V249IbGK8ml6/view?usp=sharing) | [`link`](https://pan.baidu.com/s/1FHhhGrj8qJe1bLMFeDClwQ) |
### Notes:
- The input size is 448.
- The CGNLx block with Gaussian RBF \[[0](#reference)\]\[[1](#refereExcerpt of 10,990 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:f3aafc0412ccc5d5, topic:pytorch
matched fp:f3aafc0412ccc5d5, topic:computer-vision