Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/sbbug/yolov5-prune-multi
无人机视角、多模态、模型剪枝、国产AI芯片部署
https://github.com/sbbug/yolov5-prune-multi
Last synced: 3 months ago
JSON representation
无人机视角、多模态、模型剪枝、国产AI芯片部署
- Host: GitHub
- URL: https://github.com/sbbug/yolov5-prune-multi
- Owner: sbbug
- Created: 2021-03-24T11:28:31.000Z (over 3 years ago)
- Default Branch: master
- Last Pushed: 2022-02-06T03:35:34.000Z (almost 3 years ago)
- Last Synced: 2024-06-16T14:31:50.221Z (5 months ago)
- Language: Python
- Size: 5.14 MB
- Stars: 35
- Watchers: 2
- Forks: 3
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome-yolo-object-detection - sbbug/yolov5-prune-multi - prune-multi?style=social"/> : yolov5-prune-multi 无人机视角、多模态、模型剪枝、国产AI芯片部署。 (Lighter and Deployment Frameworks)
- awesome-yolo-object-detection - sbbug/yolov5-prune-multi - prune-multi?style=social"/> : yolov5-prune-multi 无人机视角、多模态、模型剪枝、国产AI芯片部署。 (Lighter and Deployment Frameworks)
README
# yolov5-prune-multi
无人机视角、多模态、模型剪枝、国产AI芯片部署
将训练好的视觉模型,进行剪枝压缩再训练,最终部署到终端设备上。## 相关实验数据
#### 数据集展示
###### 对齐的双模态图像
![相关图片](./img/1-8.png)
###### 双模态图像类别分布
![相关图片](./img/1-7.png)#### 数据集基础实验(多模态数据集为团队收集,暂未公开)
![数据集基础实验](./img/1-6.png)
#### 多模态模型剪枝效果实验
![剪枝实验](./img/1-1.png)
#### 国产芯片与英伟达芯片对比实验
###### 硬件参数
![芯片相关参数](./img/1-5.png)
###### 对比实验(硬件性能开到最大)
![硬件性能对比](./img/1-2.png)#### 模型算法效果截图
![模型优化效果](./img/1-3.png)
#### 模型部署到边缘设备效果截图
![边缘设备部署效果](./img/1-4.png)
## 参考
[AGX板卡模型推理部署工具链](https://github.com/sbbug/yolov5-multi-tensorrt)
[地平线板卡模型推理部署工具链](https://github.com/sbbug/x3_chain)