prfans

测试程序,开启ONLY_CPU,运行不正常,结果为空;CUDNN下运行正常.

18

**ONLY_CPU结果**: I1120 15:23:28.928501 5788 net.cpp:270] This network produces output result I1120 15:23:28.928501 5788 net.cpp:283] Network initialization done. I1120 15:23:28.942502 5788 net.cpp:761] Ignoring source layer data I1120 15:23:28.946501 5788 net.cpp:761] Ignoring...

I have the same problem, all output results are zeros. Anyone can help ?

5

I have the same problem, all output results are zeros. Anyone can help ? _Originally posted by @prfans in https://github.com/wkentaro/pytorch-fcn/issues/113#issuecomment-555805034_ I used the pretrained fcn8s/fcn16s/fcn32s model, all output results are...

Input channel 1 less then 3, please update NPU Driver to v1.3.0 or later.

1

同样的模型，input channel==1转换失败，input channel==3转换成功。失败信息： D:\Users\xxxxx\Miniconda3\envs\rknn_v1.75\lib\site-packages\numpy\_distributor_init.py:32: UserWarning: loaded more than 1 DLL from .libs: D:\Users\xxxxx\Miniconda3\envs\rknn_v1.75\lib\site-packages\numpy\.libs\libopenblas.TXA6YQSD3GCQQC22GEQ54J2UDCXDXHWN.gfortran-win_amd64.dll D:\Users\xxxxx\Miniconda3\envs\rknn_v1.75\lib\site-packages\numpy\.libs\libopenblas.WCDJNK7YVMPZQ2ME2ZZHJJRJ3JIKNDB7.gfortran-win_amd64.dll stacklevel=1) W rknn-toolkit version: 1.7.5 W Detect torchvision(0.11.0). If the python program got segment fault error,...

opencl，批量推理错误

1

windows系统下，批量样本推理得不到正确的结果，例如一次推理4张图像：[4,3,256,256]结果错误，但是推理一张图像[1,3,256,256]正确，cpu下推理正确只有再opencl下推理失败，涉及的版本从2.1-2.9版本。