I try to run the demo in in Google Colab but always get this error in the "Compute Flow" step.
How can I fix this?
The full log is:
Sampled 522 frame pairs.
Resizing flow to (224, 384)
Loading pretrained model from 'checkpoints/flownet2.pth'.
error in correlation_forward_cuda_kernel: no kernel image is available for execution on the device
Traceback (most recent call last):
File "main.py", line 13, in
dp.process(params)
File "/content/consistent_depth/process.py", line 117, in process
return self.pipeline(params)
File "/content/consistent_depth/process.py", line 71, in pipeline
self.flow.compute_flow(frame_pairs, params.flow_checkpoint)
File "/content/consistent_depth/flow.py", line 143, in compute_flow
optical_flow_flownet2_homography.process(args)
File "/content/consistent_depth/optical_flow_flownet2_homography.py", line 258, in process
flow = infer(args, Flownet, device, im1, im2)
File "/content/consistent_depth/optical_flow_flownet2_homography.py", line 199, in infer
flow = Flownet(inputs)[0].permute(1, 2, 0).data.cpu().numpy()
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 532, in call
result = self.forward(*input, kwargs)
File "/content/consistent_depth/third_party/flownet2/models.py", line 124, in forward
flownetc_flow2 = self.flownetc(x)[0]
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 532, in call
result = self.forward(*input, *kwargs)
File "/content/consistent_depth/third_party/flownet2/networks/FlowNetC.py", line 86, in forward
out_corr = self.corr(out_conv3a, out_conv3b) # False
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 532, in call
result = self.forward(input, kwargs)
File "/content/consistent_depth/third_party/flownet2/networks/correlation_package/correlation.py", line 67, in forward
result = CorrelationFunction(self.pad_size, self.kernel_size, self.max_displacement,self.stride1, self.stride2, self.corr_multiply)(input1, input2)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 532, in call
result = self.forward(*input, **kwargs)
File "/content/consistent_depth/third_party/flownet2/networks/correlation_package/correlation.py", line 35, in forward
self.pad_size, self.kernel_size, self.max_displacement,self.stride1, self.stride2, self.corr_multiply)
RuntimeError: CUDA call failed (correlation_forward_cuda at correlation_cuda.cc:82)
frame #0: c10::Error::Error(c10::SourceLocation, std::string const&) + 0x33 (0x7f3908df5193 in /usr/local/lib/python3.7/dist-packages/torch/lib/libc10.so)
frame #1: correlation_forward_cuda(at::Tensor&, at::Tensor&, at::Tensor&, at::Tensor&, at::Tensor&, int, int, int, int, int, int) + 0x628 (0x7f38e9a77b78 in /root/.local/lib/python3.7/site-packages/correlation_cuda-0.0.0-py3.7-linux-x86_64.egg/correlation_cuda.cpython-37m-x86_64-linux-gnu.so)
frame #2: + 0x1bd9a (0x7f38e9a87d9a in /root/.local/lib/python3.7/site-packages/correlation_cuda-0.0.0-py3.7-linux-x86_64.egg/correlation_cuda.cpython-37m-x86_64-linux-gnu.so)
frame #3: + 0x188e0 (0x7f38e9a848e0 in /root/.local/lib/python3.7/site-packages/correlation_cuda-0.0.0-py3.7-linux-x86_64.egg/correlation_cuda.cpython-37m-x86_64-linux-gnu.so)
frame #57: __libc_start_main + 0xe7 (0x7f3962ad6bf7 in /lib/x86_64-linux-gnu/libc.so.6)
I try to run the demo in in Google Colab but always get this error in the "Compute Flow" step. How can I fix this?
The full log is: