sandy

Results 9 comments of sandy

I got same problem ..... _OpenCV Error: Assertion failed ((lutcn == cn || lutcn == 1) && lut.total() == 256 && lut.isContinuous() && (src.depth() == CV_8U || src.depth() == CV_8S))...

@wzhouuu @jinfagang for me, --colour input img should be 1 channel? can using the default _ENet/scripts/cityscapes19.png_ try

@opti-mix this is whole cmd line: ../glow/build_Debug/bin/model-compiler -load-profile=live_profile.yaml -model=./live_128.onnx \ -model-input=input,float,[1,1,128,128] -backend=CPU -target=thumbv7-linux-gnueabi -mcpu=cortex-r5 -float-abi=soft -emit-bundle=live_128_arm_r5_quanta \ -keep-original-precision-for-nodes=Div -arm-implicit-it=thumb -llvm-compiler=" -fshort-wchar" I want using live_128.o in one armcc compiled project;...

-keep-original-precision-for-nodes=Add,Sub,Mul,Div,Exp,SoftMax

> @zoufangyu1987 @kunyao2015 > 实现步骤: > 1、把模型转成caffemodel; > 2、前后处理自己c++实现; > 3、Done! > 亲测可以,祝你好运! 转caffe时候的DCNV2怎么办?caffe又不是天然支持

> 试试 https://github.com/trzsz/trzsz-go 能不能用? > > SecureCRT 新建 Session 时选择 `Local Shell`,然后选择 `cmd` 或 `PowerShell` 的路径。 在里面执行 `trzsz ssh x.x.x.x` 登录到远程服务器,就可以使用 `trz` 上传,`tsz` 下载了。 是可以的,不知道为什么速度那么慢?只是本地的虚拟机 ![image](https://user-images.githubusercontent.com/3970271/181267449-4954f526-e1be-4a10-b9ff-e0e37553c3a1.png) Receive data timeout 后就崩了