可以增加mobilenetv2么

SpursLipu / YOLOv3v4-ModelCompression-MultidatasetTraining-Multibackbone

YOLO ModelCompression MultidatasetTraining

GNU General Public License v3.0

445 stars 136 forks source link

Closed anguoyang closed 4 years ago

anguoyang commented 4 years ago

测试过人脸检测（retinaface/pytorch），mobilenetv3（即便是large）的精度不如mobilenetv2,而且差的还比较多。不知道yolov4怎么样。当然v2的确比较大

SpursLipu commented 4 years ago

你可以自己改吧mobilenet的算子，我这个仓库都支持的，你只需要重新写cfg文件就可以了

anguoyang commented 4 years ago

好的，谢谢