# Lynxi-model-zoo

**Repository Path**: lize0325/lynxi-model-zoo

## Basic Information

- **Project Name**: Lynxi-model-zoo
- **Description**: Lynxi开源modelzoo社区
- **Primary Language**: Python
- **License**: Apache-2.0
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 3
- **Created**: 2023-09-14
- **Last Updated**: 2023-09-14

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# lynxi-model-zoo

本目录主要汇集了由灵汐科技完成适配的开源深度学习模型。在各模型文件夹下，用户可以找到配套的适配、转换、测试代码，也可基于此修改和训练模型，并部署到灵汐平台。

# 环境准备

该工程需要依赖灵汐软硬件，编译转换需要Lyngor，测试需要SDK和硬件，请提前联系销售/售前获取并安装。模型在转换过程中需要安装对应的python库环境，因此强烈推荐安装virtualenv（或者conda），在python虚拟环境中进行验证。
- [x] python3.6.5
- [ ] virtualenv
- [x] lyngor
- [x] lynsdk
- [x] lyndriver

当前发布版本使用匹配软件版本如下：
- LynSDK&LynDriver V1.11.0
- Lyngor V1.11.0

``` shell
git clone https://gitee.com/Lynxi/lynxi-model-zoo.git
cd lynxi-model-zoo

# 创建并激活虚拟环境
virtualenv venv
source venv/bin/activate

# 如果下载过慢，请自行切换pypi源
pip3 install -r requirements.txt
# 获取并安装lyngor
pip3 install lyngor-x.x.x.x-cp36-cp36m-linux_x86_64.whl
```


# 模型支持列表

分类模型
- [DenseNet121](examples/opensource/caffe/cv/image_classification/DenseNet121/)
- [DenseNet201](examples/opensource/caffe/cv/image_classification/DenseNet201/)
- [Inception-v1](examples/opensource/pytorch/cv/image_classification/inceptionv1/)
- [Inception-v2](examples/opensource/caffe/cv/image_classification/inception-v2/)
- [Inception-v3](examples/opensource/caffe/cv/image_classification/inception-v3/)
- [Inception-v4](examples/opensource/caffe/cv/image_classification/inception-v4/)
- [MobileNet-v1](examples/opensource/caffe/cv/image_classification/MobileNet-v1/)
- [MobileNet-v2](examples/opensource/caffe/cv/image_classification/MobileNet-v2/)
- [Resnet18](examples/opensource/caffe/cv/image_classification/ResNet-18/)
- [Resnet50](examples/opensource/caffe/cv/image_classification/ResNet-50/)
- [SENet](examples/opensource/caffe/cv/image_classification/SENet/)
- [SqueezeNet-V1.0](examples/opensource/caffe/cv/image_classification/SqueezeNet_v1.0/)
- [SqueezeNet-V1.1](examples/opensource/caffe/cv/image_classification/SqueezeNet_v1.1/)
- [RepVGG](examples/opensource/pytorch/cv/image_classification/RepVGG/)
- [VGG16](examples/opensource/caffe/cv/image_classification/VGG/)
- [Pytorch-mobilenet-cifar100](examples/opensource/pytorch/cv/image_classification/mobilenet/)

Caffe-Yolo检测
- [yolov2](examples/opensource/caffe/cv/object_detect/yolov2/)
- [yolov3](examples/opensource/caffe/cv/object_detect/yolov3/)

分割
- [UNet-carvana](examples/opensource/pytorch/cv/segmentation/UNet/)
- [unet-voc](examples/opensource/pytorch/cv/segmentation/unet_voc/)
- [yolov5-seg](examples/opensource/pytorch/cv/segmentation/yolov5-seg/)
- [yolov8-seg](examples/opensource/pytorch/cv/segmentation/yolov8-seg/)

行人属性识别
- [Person-Attribute](examples/opensource/pytorch/cv/image_classification/Person-Attribute/)

车辆属性识别
- [Vehicle-Attribute](examples/opensource/onnx/cv/image_classification/Vehicle-Attribute/)

车牌检测
- [YOLOv5-LPRNet-Licence-Recognition](examples/opensource/pytorch/cv/object_detect/yolov5-LPRNet-Licence-Recognition/)
- [Chinese_license_plate_detection_recognition](examples/opensource/pytorch/cv/object_detect/yolov5-plate-detect/)

车牌识别
- [Caffe-plate-recognition](examples/opensource/caffe/cv/image_classification/Plate_recognition/)

烟火检测
- [yolov5-fire_smoke](examples/opensource/pytorch/cv/object_detect/yolov5-fire_smoke/)

口罩检测
- [yolov5-face_mask](examples/opensource/pytorch/cv/object_detect/yolov5-face_mask/)

人脸检测
- [yolov5-face-detect](examples/opensource/pytorch/cv/object_detect/yolov5-face-detect/)
- [yolov7-face-detect](examples/opensource/pytorch/cv/object_detect/yolov7-face-detect/)
- [Pytorch-retinaface](examples/opensource/pytorch/cv/object_detect/retinaface/)
- [insightface-scrfd](examples/opensource/onnx/cv/face_detect/scrfd/)

人脸识别
- [face_recognition_arcface](examples/opensource/onnx/cv/face_recognition/arcface/)

人脸性别年龄
- [face_attribute](examples/opensource/onnx/cv/face_attribute/insightface)

通用目标检测
- [Yolov5](examples/opensource/pytorch/cv/object_detect/ultralytics_yolov5/)
- [yolov8](examples/opensource/pytorch/cv/object_detect/yolov8/)
- [Pytorch-SSD](examples/opensource/pytorch/cv/object_detect/ssd/)
- [Pytorch-FCOS](examples/opensource/pytorch/cv/object_detect/FCOS/)
- [Pytorch-YOLOX](examples/opensource/pytorch/cv/object_detect/Megvii_YOLOX/)

人体关键点检测
- [openpose-pytorch](examples/opensource/pytorch/cv/object_detect/openpose/)
- [yolov8-pose](examples/opensource/pytorch/cv/object_detect/yolov8-pose/)

文本检测识别(文本检测+文字方向预测+文字识别)
- [paddlepaddle-OCRv3](examples/opensource/onnx/cv/ocr/paddlepaddle/)

目标跟踪
- [DeepSort](examples/opensource/pytorch/cv/tracking/yolov5-deepsort/)

交通流量
- [STSGCN](examples/opensource/pytorch/tsf/trafficflow/stsgcn/)

## 免责声明

您明确了解并同意，本例程中的软件、数据或者模型由第三方提供并负责维护。在本例程中出现的任何第三方的名称、商标、标识、产品或服务并不构成明示或暗示与该第三方或其软件、数据或模型的相关背书、担保或推荐行为。您进一步了解并同意，使用任何第三方软件、数据或者模型，包括您提供的任何信息或个人数据（不论是有意或无意地），应受相关使用条款、许可协议、隐私政策或其他此类协议的约束。因此，使用本例程中的软件、数据或者模型可能导致的所有风险将由您自行承担。