Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

no module named timm #150

Open
wants to merge 11 commits into
base: iscyy-beta
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -3,4 +3,5 @@
yolov5s.pt
log_model
runs
*.idea
*.idea
coco128
94 changes: 62 additions & 32 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,3 +1,10 @@
# YOLOAir - yolov7分支
🌟🌟🌟YOLOv7, YOLOv5, Transformer, YOLOR, YOLOX and other networks... Support to improve backbone, head, loss, IoU, NMS, Rep and other modules, based on the [YOLOv7](https://github.com/WongKinYiu/yolov7) framework

此分支以 YOLOv7 改进为核心,

______________________________________________________________________

## YOLO Air: Make the improvement of the YOLO model faster, more convenient

<div align="center">
Expand All @@ -6,7 +13,7 @@
<p>统一模型代码框架、统一应用、统一改进、易于模块组合、构建更强大的网络模型。</p>
<p>
<a align="center" href="https://github.com/iscyy/yoloair" target="_blank">
<img width="850" src="docs/image/logo00.png"></a>
<img width="850" src="https://github.com/iscyy/yoloair/raw/main/docs/image/logo00.png"></a>
<br><br>
</p>

Expand All @@ -19,7 +26,7 @@
![](https://img.shields.io/badge/Support-YOLOv5-red) ![](https://img.shields.io/badge/Support-YOLOv7-brightgreen) ![](https://img.shields.io/badge/Support-YOLOX-yellow) ![](https://img.shields.io/badge/Support-YOLOv4-green) ![](https://img.shields.io/badge/Support-Scaled_YOLOv4-ff96b4)
![](https://img.shields.io/badge/Support-YOLOv3-yellowgreen) ![](https://img.shields.io/badge/Support-YOLOR-lightgrey) ![](https://img.shields.io/badge/Support-Transformer-9cf) ![](https://img.shields.io/badge/Support-Attention-green)

[特性🚀](#主要特性) • [使用🍉](#使用) • [文档📒](https://github.com/iscyy/yoloair) • [报告问题🌟](https://github.com/iscyy/yoloair/issues/new) • [更新💪](#更新-)
[特性🚀](#主要特性) • [使用🍉](#使用) • [文档📒](https://github.com/iscyy/yoloair) • [报告问题🌟](https://github.com/iscyy/yoloair/issues/new) • [更新💪](#-to-do不定时更新)

</div>

Expand All @@ -30,7 +37,7 @@
**统一模型代码框架、统一应用方式、统一调参、统一改进、易于模块组合、构建更强大的网络模型**。


内置YOLOv5、YOLOv7、YOLOX、YOLOR、Transformer、Scaled_YOLOv4、YOLOv3、YOLOv4、YOLO-Facev2、TPH-YOLOv5、YOLOv5Lite、SPD-YOLO、PicoDet等模型网络结构(持续更新中🚀)...
内置YOLOv5、YOLOv7、YOLOX、YOLOR、Transformer、Scaled_YOLOv4、YOLOv3、YOLOv4、YOLO-Facev2、TPH-YOLO、YOLOv5Lite、SPD-YOLO、SlimNeck-YOLO、PicoDet等模型网络结构(持续更新中🚀)...

```

Expand All @@ -44,27 +51,40 @@

基于 YOLOv5 代码框架,并同步适配 **稳定的YOLOv5_v6.1更新**, 同步v6.1部署生态。使用这个项目之前, 您可以先了解YOLOv5库。

点 **Star🌟、Fork** 不迷路,同步更新。

项目地址🌟: https://github.com/iscyy/yoloair


部分改进说明教程🌟: [https://github.com/iscyy/yoloair/wiki/Improved-tutorial-presentation](https://github.com/iscyy/yoloair/wiki/Improved-tutorial-presentation)

______________________________________________________________________

### 🌟 To Do(不定时更新)
下列功能 YOLOAir-Beta版本内测 已支持🔥

- 已经完成更新 **20+ 种 Transformer系列及其变种主干网络**🌟、**多种MLP网络🌟** 以及 **大部分重参数化结构模型网络** 🚀🚀🚀
- 已经完成更新 **图神经网络**🌟 在YOLO中的应用
- 已经完成更新 多模态🔥 在YOLO中的应用
- 已经完成更新 30+ 种 **Attention注意力机制**🌟、
- 已经完成更新 **多种Head检测头**
- 已经完成更新 **YOLO系列网络模型热力图可视化**(GardCAM、GardCAM++等)支持YOLOv3、YOLOv3-SPP、YOLOv4、YOLOv5、YOLOR、YOLOv7、Scaled_YOLOv4、TPH-YOLO、SPD-YOLO以及自定义网络模型等模型
- 其他 各种
______________________________________________________________________

### 主要特性🚀

🚀支持更多的YOLO系列算法模型改进(持续更新...)

YOLOAir 算法库汇总了多种主流YOLO系列检测模型,一套代码集成多种模型:
- 内置集成 YOLOv5 模型网络结构、YOLOv7 模型网络结构、 YOLOR 模型网络结构、YOLOX 模型网络结构、Scaled_YOLOv4 模型网络结构、YOLOv4 模型网络结构、YOLOv3 模型网络结构、YOLO-FaceV2模型网络结构、TPH-YOLOv5模型网络结构、SPD-YOLO模型网络结构、YOLOv5-Lite模型网络结构、PicoDet模型网络结构等持续更新中...
- 内置集成 YOLOv5 模型网络结构、YOLOv7 模型网络结构、 YOLOR 模型网络结构、YOLOX 模型网络结构、Scaled_YOLOv4 模型网络结构、YOLOv4 模型网络结构、YOLOv3 模型网络结构、YOLO-FaceV2模型网络结构、TPH-YOLOv5模型网络结构、SPD-YOLO模型网络结构、SlimNeck-YOLO模型网络结构、YOLOv5-Lite模型网络结构、PicoDet模型网络结构等持续更新中...

|||
:-------------------------:|:-------------------------:
<img src='docs/image/test.jpg' width='500px' alt='ingishvcn'> | <img src='docs/image/zebra.jpg' width='500px' alt='ingishvcn'>
<img src='https://github.com/iscyy/yoloair/raw/main/docs/image/test.jpg' width='500px' alt='ingishvcn'> | <img src='https://github.com/iscyy/yoloair/raw/main/docs/image/zebra.jpg' width='500px' alt='ingishvcn'>

- 以上多种检测算法使用统一模型代码框架,**集成在 YOLOAir 库中,统一任务形式、统一应用方式**。🌟便于科研者用于论文算法模型改进,模型对比,实现网络组合多样化。🌟工程算法部署落地更便捷,包含轻量化模型和精度更高的模型,根据场景合理选择,在精度和速度俩个方面取得平衡。同时该库支持解耦不同的结构和模块组件,让模块组件化,通过组合不同的模块组件,用户可以根据不同数据集或不同业务场景自行定制化构建不同检测模型。

🔥🔥🔥 重磅!!!作为开源项目补充,推荐一个注意力机制算法代码库[External-Attention](https://github.com/xmu-xiaoma666/External-Attention-pytorch),里面汇总整理很全面,包含各种Attention等代码,代码简洁易读,一行代码实现Attention机制。

🚀支持加载YOLOv3、YOLOv4、YOLOv5、YOLOv7、YOLOR、Scaled_YOLO等网络的官方预训练权重进行迁移学习

🚀支持更多Backbone
Expand All @@ -81,6 +101,7 @@ YOLOAir 算法库汇总了多种主流YOLO系列检测模型,一套代码集
- RepLKNet系列
- 重参数化系列
- RepVGG系列
- RepMLP系列
- ACNet系列
- RepConv系列
- Mobileone系列
Expand Down Expand Up @@ -142,6 +163,7 @@ YOLOAir 算法库汇总了多种主流YOLO系列检测模型,一套代码集
🚀支持更多Loss
- ComputeLoss
- ComputeLoss(v5)
- ComputeLoss(v6)
- ComputeLoss(X)
- ComputeLossAuxOTA(v7)
- ComputeLossOTA(v7)
Expand Down Expand Up @@ -188,36 +210,33 @@ Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, SPPF, DWConv, MixConv2d, Focu

以上组件模块使用统一模型代码框架、统一任务形式、统一应用方式,**模块组件化**🚀 可以帮助用户自定义快速组合Backbone、Neck、Head,使得网络模型多样化,助力科研改进检测算法,构建更强大的网络模型。

______________________________________________________________________

### 内置网络模型配置支持✨

🚀包括基于 YOLOv5、YOLOv7、YOLOX、YOLOR、YOLOv3、YOLOv4、Scaled_YOLOv4、Transformer、YOLO-FaceV2、PicoDet、YOLOv5-Lite、TPH-YOLOv5、SPD-YOLO等**其他多种改进网络结构等算法模型**的模型配置文件
______________________________________________________________________

### 更新 🌟

🔥添加YOLOv3、YOLOv4、YOLOv5、YOLOv7、Scaled-YOLOv4网络的不同预训练权重
持续更新中🎈
______________________________________________________________________

### 预训练权重🚀

- YOLOv5
https://github.com/ultralytics/yolov5/releases/tag/v6.1

- YOLOv4
https://github.com/iscyy/yoloair/releases/download/v1.0
https://github.com/iscyy/yoloair/releases/tag/v1.0

- YOLOv3
https://github.com/iscyy/yoloair/releases/download/v1.0
https://github.com/iscyy/yoloair/releases/tag/v1.0

- YOLOR
https://github.com/iscyy/yoloair/releases/download/v1.0
https://github.com/iscyy/yoloair/releases/tag/v1.0

- Scaled_YOLO
https://github.com/iscyy/yoloair/releases/download/v1.0
https://github.com/iscyy/yoloair/releases/tag/v1.0

- YOLOv7
https://github.com/iscyy/yoloair/releases/download/v1.0
https://github.com/iscyy/yoloair/releases/tag/v1.0

______________________________________________________________________

Expand Down Expand Up @@ -262,18 +281,20 @@ $ python wbf.py
```
______________________________________________________________________

### 技术交流 <img title="" src="https://user-images.githubusercontent.com/48054808/157800467-2a9946ad-30d1-49a9-b9db-ba33413d9c90.png" alt="" width="20">

|FightingCV公众号|YOLOAir目标检测交流群( 答案: yoloair )|
:-------------------------:|:-------------------------:
<img src='https://p3-juejin.byteimg.com/tos-cn-i-k3u1fbpfcp/b6f5057da9a8410fa22dcc7566548193~tplv-k3u1fbpfcp-watermark.image?' width='200px'> | <img src='https://p1-juejin.byteimg.com/tos-cn-i-k3u1fbpfcp/1882e8cf3f804918a043d70de1b70e31~tplv-k3u1fbpfcp-watermark.image' width='200px'>

- FightingCV每天分享前沿论文动态(公众号回复加群, 添加小助手, 加入微信交流群)

- YOLOAir目标检测交流群
______________________________________________________________________

### Performance
#### YOLOv7
| Model | Test Size | AP<sup>test</sup> | AP<sub>50</sub><sup>test</sup> | AP<sub>75</sub><sup>test</sup> | batch 1 fps | batch 32 average time |
| :-- | :-: | :-: | :-: | :-: | :-: | :-: |
| [**YOLOv7**](https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7.pt) | 640 | **51.4%** | **69.7%** | **55.9%** | 161 *fps* | 2.8 *ms* |
| [**YOLOv7-X**](https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7x.pt) | 640 | **53.1%** | **71.2%** | **57.8%** | 114 *fps* | 4.3 *ms* |
| | | | | | | |
| [**YOLOv7-W6**](https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7-w6.pt) | 1280 | **54.9%** | **72.6%** | **60.1%** | 84 *fps* | 7.6 *ms* |
| [**YOLOv7-E6**](https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7-e6.pt) | 1280 | **56.0%** | **73.5%** | **61.2%** | 56 *fps* | 12.3 *ms* |
| [**YOLOv7-D6**](https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7-d6.pt) | 1280 | **56.6%** | **74.0%** | **61.8%** | 44 *fps* | 15.0 *ms* |
| [**YOLOv7-E6E**](https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7-e6e.pt) | 1280 | **56.8%** | **74.4%** | **62.1%** | 36 *fps* | 18.7 *ms* |

#### YOLOv5

| Model | size<br><sup>(pixels) | mAP<sup>val<br>0.5:0.95 | mAP<sup>val<br>0.5 | Speed<br><sup>CPU b1<br>(ms) | Speed<br><sup>V100 b1<br>(ms) | Speed<br><sup>V100 b32<br>(ms) | params<br><sup>(M) | FLOPs<br><sup>@640 (B) | Weights
|------------------------------------------------------------------------------------------------------|-----------------------|-------------------------|--------------------|------------------------------|-------------------------------|--------------------------------|--------------------|------------------------|------------------------|
| YOLOv5n | 640 | 28.0 | 45.7 | **45** | **6.3** | **0.6** | **1.9** | **4.5** | [YOLOv5n](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5n.pt)
Expand Down Expand Up @@ -377,6 +398,17 @@ ______________________________________________________________________

______________________________________________________________________

## Citation✨

```python
@article{2022yoloair,
title={{YOLOAir}: Make the improvement of the YOLO model faster, more convenient},
author={iscyy},
repo={github https://github.com/iscyy/yoloair},
year={2022}
}
```

## Statement
<details><summary> <b>Expand</b> </summary>

Expand All @@ -390,7 +422,6 @@ ______________________________________________________________________
<details><summary> <b>Expand</b> </summary>

[https://github.com/ultralytics/yolov5](https://github.com/ultralytics/yolov5)
[https://github.com/AlexeyAB/darknet](https://github.com/AlexeyAB/darknet)
[https://github.com/ultralytics/yolov3](https://github.com/ultralytics/yolov3)
[https://github.com/WongKinYiu/PyTorch_YOLOv4](https://github.com/WongKinYiu/PyTorch_YOLOv4)
[https://github.com/WongKinYiu/ScaledYOLOv4](https://github.com/WongKinYiu/ScaledYOLOv4)
Expand All @@ -402,8 +433,7 @@ ______________________________________________________________________
[https://github.com/Krasjet-Yu/YOLO-FaceV2](https://github.com/Krasjet-Yu/YOLO-FaceV2)
[https://github.com/positive666/yolov5_research/](https://github.com/positive666/yolov5_research)
[https://github.com/ppogg/YOLOv5-Lite](https://github.com/ppogg/YOLOv5-Lite)
[https://github.com/Gumpest/YOLOv5-Multibackbone-Compression](https://github.com/Gumpest/YOLOv5-Multibackbone-Compression)
[https://github.com/Gumpest/YOLOv5-Multibackbone](https://github.com/Gumpest/YOLOv5-Multibackbone-Compression)
[https://github.com/cv516Buaa/tph-yolov5](https://github.com/cv516Buaa/tph-yolov5)
Paper:[https://arxiv.org/abs/2208.02019](https://arxiv.org/abs/2208.02019)

</details>
</details>
Loading