labdeeman7
commented
4 years ago I have a similar error as well with MPI Sintel dataset, when I visualize the flo file, I get blurred images.
I also have the same warning when running install.sh and the same error when I perform inference : error in correlation_forward_cuda_kernel: CUDA driver version is insufficient for CUDA runtime version
My environment is colab Tesla K40c GPU Cuda 9.0.176 py 3.6 pytorch 1.0.0 cuda 90 gcc-6 v6.1 torchvision 0.2.1
Any help would be appreciated
coly-ai
wgh-2018
nudlesoup
Hi I am running inference on the flying chair Dataset and not able to get expected results : These are my envs : Env 1 on conda : Linux Ubuntu - 14.4.6 Tesla K40c GPU Cuda 9.0.176 py 3.6 pytorch 1.0.0 - py3.6_cuda9.0.176_cudnn7.4.1_1 cuda 90 gcc-6 v6.1 torchvision 0.2.2 When i run install.sh i get these warnings but no errors.
gcc -pthread -B /home//miniconda3/envs/test1/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/lib/include -I/home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/lib/include/TH -I/home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/home/xyz/miniconda3/envs/test1/include/python3.6m -c correlation_cuda.cc -o build/temp.linux-x86_64-3.6/correlation_cuda.o -std=c++11 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=correlation_cuda -D_GLIBCXX_USE_CXX11_ABI=0 cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++ In file included from correlation_cuda.cc:1:0: /home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include/torch/torch.h:7:2: warning: #warning "Including torch/torch.h for C++ extensions is deprecated. Please include torch/extension.h" [-Wcpp]
warning \
^
~~ correlation_cuda.cc: In function ‘int correlation_forward_cuda(at::Tensor&, at::Tensor&, at::Tensor&, at::Tensor&, at::Tensor&, int, int, int, int, int, int)’: correlation_cuda.cc:44:7: warning: unused variable ‘success’ [-Wunused-variable] int success = correlation_forward_cuda_kernel( ^~~ /usr/local/cuda/bin/nvcc -I/home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/lib/include -I/home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/lib/include/TH -I/home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/home/xyz/miniconda3/envs/test1/include/python3.6m -c correlation_cuda_kernel.cu -o build/temp.linux-x86_64-3.6/correlation_cuda_kernel.o -DCUDA_NO_HALF_OPERATORS -DCUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_HALF2_OPERATORS__ --compiler-options '-fPIC' -gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_70,code=compute_70 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=correlation_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++11 creating build/lib.linux-x86_64-3.6I am able to run inference but get the following warning and poor result.
error in correlation_forward_cuda_kernel: CUDA driver version is insufficient for CUDA runtime /home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/nn/modules/upsampling.py:129: UserWarning: nn.Upsample is deprecated. Use nn.functional.interpolate instead. warnings.warn("nn.{} is deprecated. Use nn.functional.interpolate instead.".format(self.name)) /home/xyz/miniconda3/envs/test1/lib/python3.6/site-packages/torch/nn/functional.py:2423: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details. "See the documentation of nn.Upsample for details.".format(mode)) This is my results
while expected result is :
My 2nd env has the following but still gives similar results as above : cuda 9 gcc-5 5.4 colorama 0.4.3 py_0 ld_impl_linux-64 2.33.1 h53a641e_7 libedit 3.1.20181209 hc058e9b_0 libffi 3.3 he6710b0_1 libgcc-ng 9.1.0 hdf63c60_0 libgfortran-ng 7.3.0 hdf63c60_0 matplotlib 3.2.1 pypi_0 pypi protobuf 3.12.2 pypi_0 pypi python 3.6.10 h7579374_2 python-dateutil 2.8.1 pypi_0 pypi pytorch 0.4.1 py36_py35_py27__9.0.176_7.1.2_2 pytorch pytz 2020.1 pypi_0 pypi readline 8.0 h7b6447c_0 scipy 1.1.0 pypi_0 pypi setproctitle 1.1.10 pypi_0 pypi tensorboardx 2.0 pypi_0 pypi tk 8.6.8 hbc83047_0 torchvision 0.2.1 py_2 pytorch
I get same poor results as above. But during inference i get this error : error in correlation_forward_cuda_kernel: no kernel image is available for execution on the device /home/xyz/anaconda3/envs/flow/lib/python3.6/site-packages/torch/nn/modules/upsampling.py:129: UserWarning: nn.Upsample is deprecated. Use nn.functional.interpolate instead. warnings.warn("nn.{} is deprecated. Use nn.functional.interpolate instead.".format(self.name)) /home/xyz/anaconda3/envs/flow/lib/python3.6/site-packages/torch/nn/functional.py:2423: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details. "See the documentation of nn.Upsample for details.".format(mode))
Can someone please guide me on how to fix this? Any ideas would be much appreciated.