YOLOX README

The project's README from github.com/Megvii-BaseDetection/YOLOX. As one markdown file: README.md.

Introduction

YOLOX is an anchor-free version of YOLO, with a simpler design but better performance! It aims to bridge the gap between research and industrial communities. For more details, please refer to our report on Arxiv.

This repo is an implementation of PyTorch version YOLOX, there is also a MegEngine implementation.

Updates!!

Benchmark

Standard Models.

Model size mAPval
0.5:0.95		 mAPtest
0.5:0.95		 Speed V100
(ms)		 Params
(M)		 FLOPs
(G)		 weights
YOLOX-s 640 40.5 40.5 9.8 9.0 26.8 github
YOLOX-m 640 46.9 47.2 12.3 25.3 73.8 github
YOLOX-l 640 49.7 50.1 14.5 54.2 155.6 github
YOLOX-x 640 51.1 51.5 17.3 99.1 281.9 github
YOLOX-Darknet53 640 47.7 48.0 11.1 63.7 185.3 github
Legacy models
Model size mAPtest
0.5:0.95		 Speed V100
(ms)		 Params
(M)		 FLOPs
(G)		 weights
YOLOX-s 640 39.6 9.8 9.0 26.8 onedrive/github
YOLOX-m 640 46.4 12.3 25.3 73.8 onedrive/github
YOLOX-l 640 50.0 14.5 54.2 155.6 onedrive/github
YOLOX-x 640 51.2 17.3 99.1 281.9 onedrive/github
YOLOX-Darknet53 640 47.4 11.1 63.7 185.3 onedrive/github

Light Models.

Model size mAPval
0.5:0.95		 Params
(M)		 FLOPs
(G)		 weights
YOLOX-Nano 416 25.8 0.91 1.08 github
YOLOX-Tiny 416 32.8 5.06 6.45 github
Legacy models
Model size mAPval
0.5:0.95		 Params
(M)		 FLOPs
(G)		 weights
YOLOX-Nano 416 25.3 0.91 1.08 github
YOLOX-Tiny 416 32.8 5.06 6.45 github

Quick Start

Installation

Step1. Install YOLOX from source.

git clone [email protected]:Megvii-BaseDetection/YOLOX.git
cd YOLOX
pip3 install -v -e .  # or  python3 setup.py develop
Demo

Step1. Download a pretrained model from the benchmark table.

Step2. Use either -n or -f to specify your detector's config. For example:

python tools/demo.py image -n yolox-s -c /path/to/your/yolox_s.pth --path assets/dog.jpg --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]

or

python tools/demo.py image -f exps/default/yolox_s.py -c /path/to/your/yolox_s.pth --path assets/dog.jpg --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]

Demo for video:

python tools/demo.py video -n yolox-s -c /path/to/your/yolox_s.pth --path /path/to/your/video --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]
Reproduce our results on COCO

Step1. Prepare COCO dataset

cd <YOLOX_HOME>
ln -s /path/to/your/COCO ./datasets/COCO

Step2. Reproduce our results on COCO by specifying -n:

python -m yolox.tools.train -n yolox-s -d 8 -b 64 --fp16 -o [--cache]
                               yolox-m
                               yolox-l
                               yolox-x
  • -d: number of gpu devices
  • -b: total batch size, the recommended number for -b is num-gpu * 8
  • --fp16: mixed precision training
  • --cache: caching imgs into RAM to accelarate training, which need large system RAM.

When using -f, the above commands are equivalent to:

python -m yolox.tools.train -f exps/default/yolox_s.py -d 8 -b 64 --fp16 -o [--cache]
                               exps/default/yolox_m.py
                               exps/default/yolox_l.py
                               exps/default/yolox_x.py

Multi Machine Training

We also support multi-nodes training. Just add the following args:

  • --num_machines: num of your total training nodes
  • --machine_rank: specify the rank of each node

Suppose you want to train YOLOX on 2 machines, and your master machines's IP is 123.123.123.123, use port 12312 and TCP.

On master machine, run

python tools/train.py -n yolox-s -b 128 --dist-url tcp://123.123.123.123:12312 --num_machines 2 --machine_rank 0

On the second machine, run

python tools/train.py -n yolox-s -b 128 --dist-url tcp://123.123.123.123:12312 --num_machines 2 --machine_rank 1

Logging to Weights & Biases

To log metrics, predictions and model checkpoints to W&B use the command line argument --logger wandb and use the prefix "wandb-" to specify arguments for initializing the wandb run.

python tools/train.py -n yolox-s -d 8 -b 64 --fp16 -o [--cache] --logger wandb wandb-project <project name>
                         yolox-m
                         yolox-l
                         yolox-x

An example wandb dashboard is available here

Others

See more information with the following command:

python -m yolox.tools.train --help
Evaluation

We support batch testing for fast evaluation:

python -m yolox.tools.eval -n  yolox-s -c yolox_s.pth -b 64 -d 8 --conf 0.001 [--fp16] [--fuse]
                               yolox-m
                               yolox-l
                               yolox-x
  • --fuse: fuse conv and bn
  • -d: number of GPUs used for evaluation. DEFAULT: All GPUs available will be used.
  • -b: total batch size across on all GPUs

To reproduce speed test, we use the following command:

python -m yolox.tools.eval -n  yolox-s -c yolox_s.pth -b 1 -d 1 --conf 0.001 --fp16 --fuse
                               yolox-m
                               yolox-l
                               yolox-x
Tutorials

Deployment

  1. MegEngine in C++ and Python
  2. ONNX export and an ONNXRuntime
  3. TensorRT in C++ and Python
  4. ncnn in C++ and Java
  5. OpenVINO in C++ and Python
  6. Accelerate YOLOX inference with nebullvm in Python

Third-party resources

Cite YOLOX

If you use YOLOX in your research, please cite our work by using the following BibTeX entry:

 @article{yolox2021,
  title={YOLOX: Exceeding YOLO Series in 2021},
  author={Ge, Zheng and Liu, Songtao and Wang, Feng and Li, Zeming and Sun, Jian},
  journal={arXiv preprint arXiv:2107.08430},
  year={2021}
}

In memory of Dr. Jian Sun

Without the guidance of Dr. Jian Sun, YOLOX would not have been released and open sourced to the community. The passing away of Dr. Sun is a huge loss to the Computer Vision field. We add this section here to express our remembrance and condolences to our captain Dr. Sun. It is hoped that every AI practitioner in the world will stick to the belief of "continuous innovation to expand cognitive boundaries, and extraordinary technology to achieve product value" and move forward all the way.

没有孙剑博士的指导，YOLOX也不会问世并开源给社区使用。 孙剑博士的离去是CV领域的一大损失，我们在此特别添加了这个部分来表达对我们的“船长”孙老师的纪念和哀思。 希望世界上的每个AI从业者秉持着“持续创新拓展认知边界，非凡科技成就产品价值”的观念，一路向前。

YOLOX (Megvii-BaseDetection/YOLOX)

Searching Database…

10.7K ·GitHub
YOLOX
What is this?YOLOX is an object detector: an anchor-free version of YOLO with a simpler design but better performance, built to bridge the gap between research and industrial communities. The project says it outperforms YOLOv3-v5.
Is it a ready-trained model, a library, or training code?All three: a PyTorch implementation of YOLOX that ships pretrained weights for every model size plus training and evaluation code. A separate MegEngine implementation exists as well.
How do I get it running?Install YOLOX from source, download a pretrained model from the benchmark table, then run the demo, naming the model with -n or pointing to its config file with -f. A video demo works the same way.
What do I need to reproduce the COCO benchmark numbers?The COCO dataset, prepared before launching a training run with the -n flag. If one machine is not enough, multi-node training across machines is supported.
Is training fast, and does it need much RAM?Training was optimized to run 2x faster with about 1% higher performance. The optional --cache flag speeds it up further by caching images into RAM, which needs large system RAM.
Can I try it in a browser before installing anything?Yes. YOLOX-s and YOLOX-Nano are integrated into ModelScope with online demos, and there is a Gradio web demo on Huggingface Spaces.
What ready-made integrations exist beyond this repo?Community projects include ncnn-android-yolox with video support on Android, YOLOX-ROS for ROS2 Foxy, DeepStream deployment, Tengine support, MNN/TNN/ONNXRuntime C++ ports, and StreamYOLO for streaming perception.
Does training integrate with experiment trackers?Yes — pass --logger wandb to log metrics, predictions and model checkpoints to Weights & Biases, with wandb- prefixed arguments for initializing the run.
Why does the README end with a tribute to Dr. Jian Sun?Dr. Jian Sun guided YOLOX and its open-source release; after his passing the team added the section in his memory, calling his death a huge loss to the computer vision field.
What are similar projects?yolov3, RelateAnything, yolov5, yolov7, High-Angle_Robust_Fast_FaceAlignment, PaddleDetection, yolov10, ncnn
Which topics is it in?openvino, ncnn, tensorrt, yolo, object-detection, pretrained-models, onnx, training, deployment, neural-network, inference, computer-vision
Creator
Megvii-BaseDetection
Size
7.5 MB
Stars
10.7K
Updated
CommitJun 08Release0.3.0Apr 22
Age
5y
Commits · 90 Days