KeyError: 'MkldnnConvolutionBackward'

[maliho0803]

hi, I just run your code, appeared this problem. what is wrong with the code?

[dzyjjpy]

what's your version for pytorch?

[maliho0803]

'1.0.0.dev20190409'

在 2019年4月11日，下午12:09，jpy.cas [email protected] 写道：

what's your version for pytorch?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/longcw/pytorch2caffe/issues/25#issuecomment-481957579, or mute the thread https://github.com/notifications/unsubscribe-auth/AOVptJs2KjHdJOxutM00gOQCVJF-Q1z_ks5vfrVfgaJpZM4cmcUU.

[dzyjjpy]

met similar issue when using 1.0.0. pls use torch 0.2.0 to have a try

[791188299]

@maliho0803 Has the problem been solved? Can you share your method? Thank you very much.

[you-old]

？？

[jianfeifeng]

I also met this problem when i try to convert inception_v3 from pytorch to caffe.Has anyone solved this problem already? would you like to share the solution with us ? Thank you !

[joehishi]

sudo pip2 install http://download.pytorch.org/whl/cu80/torch-0.2.0.post3-cp27-cp27mu-manylinux1_x86_64.whl sudo pip2 install torchvision==0.1.8

[twirlined]

Met the same problem while trying to convert pytorch to caffe with pytorch 1.1.0 and torchvision 0.3.0. Does changing pytorch and torchvision version solve this problem? Thank you.

[arya-none]

when use 0.2.0, there's a new error: output_var = model(input_var) File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/nn/modules/module.py", line 224, in call result = self.forward(*input, **kwargs) File "/usr/local/python3.7.5/lib/python3.7/site-packages/torchvision/models/resnet.py", line 139, in forward x = self.conv1(x) File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/nn/modules/module.py", line 224, in call result = self.forward(*input, **kwargs) File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 254, in forward self.padding, self.dilation, self.groups) File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/nn/functional.py", line 52, in conv2d return f(input, weight, bias) TypeError: argument 0 is not a Variable