Eric Liu

icon company Avisonic icon location Hsinchu,Taiwan icon location An algorithm engineer in computer vision

Results 85 comments of Eric Liu

mobilenetv2_yolov3 caffe2ncnn转出的模型不能运行

其實改成一般的convolution即可，當group = channel 時，Convolution , DepthWiseConv 與 ConvolutionDepthWise 這三種輸出的結果基本沒有差別或差異小到可以忽視，只差在Caffe traing 和 inference的速度。

Pruning code for Mobilnet-yolov3_lite

Author github : https://github.com/lusenkong/Caffemodel_Compress Code was put here : https://github.com/eric612/MobileNet-YOLO/blob/master/src/caffe/Pruner.cpp And entry points was here : https://github.com/eric612/MobileNet-YOLO/blob/master/tools/channel_pruner.cpp Modify parameter here (default m2-yolov3) : https://github.com/eric612/MobileNet-YOLO/blob/master/sys_test_config.xml Usage : ``` cd caffe_root ./tools/channel_pruner...

Pruning code for Mobilnet-yolov3_lite

Thanks for your kindly remind , I forgot I change the anchors in my local repo , I already update the biases , but I need a few days to...

Pruning code for Mobilnet-yolov3_lite

Yes , because the code can't automatically modify the group number , but it could not make error from original author's model , he use the different layer "ConvolutionDepthwise", which...

Pruning code for Mobilnet-yolov3_lite

Thanks for your contribution

Pruning code for Mobilnet-yolov3_lite

You can see this issue : https://github.com/eric612/MobileNet-YOLO/issues/92

Pruning code for Mobilnet-yolov3_lite

@zpge cd into caffe root ``` cmake . make -j4 ```

pretrained model

You can see this issue : https://github.com/eric612/MobileNet-YOLO/issues/161 Howerver , here has a problem , I can't bring the pytorch fc layer bias to caffemodel , eval in the imagenet dataset...

mobilenet_v3_small

我轉完沒有訓練過，連結如下： https://drive.google.com/drive/folders/1jdvM4lJ6CNgySFf8_aRUmvL9mU3m9N6p?usp=sharing

mobilenet_v3_small

好的，如果你有訓練好的模型，建議可以開個pr上來