zhujian
> 碰到了同样的问题，请问解决了吗 还没，项目暂停了，搞其他事情去了
> 参考交流群意见，重新编译转换器，添加script支持 ``` cmake .. -DMNN_BUILD_CONVERTER=true -DMNN_BUILD_TORCH=ON && make -j4 ``` 重新运行，出现了其他错误： ``` $ ./MNNConvert -f TORCH --modelFile model_script.pt --MNNModel model_script.mnn --bizCode biz Start to Convert Other Model Format To...
> 下载最新仓库源码，重新进行编译并开启TORCH支持开关，能够完成AlexNet模型的转换
> 你好，看到你提到交流群，请问我如何加入交流群呢？MNN项目里的钉钉二维码都失效了。 非常感谢～～ https://github.com/alibaba/MNN#how-to-discuss-and-get-help-from-mnn-community https://github.com/alibaba/MNN/blob/master/README_CN.md#%E7%A4%BE%E5%8C%BA%E4%BA%A4%E6%B5%81%E4%B8%8E%E5%8F%8D%E9%A6%88
> 看看这个#1605，会不会对你有帮助
hi @zhanghang1989 @FrancescoSaverioZuppichini. i modified code to match the same results compare with origin code. ## code ``` # -*- coding: utf-8 -*- """ @date: 2021/1/4 上午11:32 @file: split_attention_conv2d.py @author:...
> Hi, > > I'm new to this fantastic work. When I tried to test the inference accuracy of resnet-18 with r18_torchvision_imagenet_224.yaml/r18_zcls_imagenet_224.yaml, the result accuracy is 68.136%, which is lower...
> Sorry for the slow response. > > The easiest way to do this would be to create a normal resnet34 (without masks), and then copy across the weights. >...
> @caikengxiaonengshou 能否具体解释一下开根号前后的差异性？另外有相关的试验结果木有？
> @caikengxiaonengshou 嗯嗯 我先研究一下，相关的实验估计也得重做。欢迎PR