InvalidArgumentError (see above for traceback): Restoring from checkpoint failed. This is most likely due to a mismatch between the current graph and the graph from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error: No OpKernel was registered to support Op 'Correlation' with these attrs. Registered devices: [CPU], Registered kernels: device='GPU' [[Node: FlowNet2/FlowNetCSS/FlowNetCS/FlowNetC/Correlation = Correlation[kernel_size=1, max_displacement=20, pad=20, stride_1=1, stride_2=2](FlowNet2/FlowNetCSS/FlowNetCS/FlowNetC/conv3/lrelu/add, FlowNet2/FlowNetCSS/FlowNetCS/FlowNetC/conv3_1/lrelu/add)]]

sampepose / flownet2-tf

FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks

MIT License

403 stars 195 forks source link

Closed mengyaaa closed 5 years ago

Iamanorange commented 5 years ago

Should run Flownet2 on GPU only.

fighttiger25 commented 5 years ago

Should run Flownet2 on GPU only.

Sorry, I meet the same problem. Could you please explain how to fix this problem in details, i.e. how to run Flownet2 on GPU only?

My env is Python2.7, cuda 10, tensorflow 1.12.0, without install cudnn, on Ubutun 18.

BTW, I can run flownet_s, flonet_sd, but fail with the others.

pushkalkatara commented 4 years ago

@lamanorange I'm running on GPU only, still facing the same problem.