deform_conv_cuda ImportError

[starry-xin]

Is there a requirement for the CUDA version?
ImportError happens when I run the demo：

$ ./scripts/inference.sh configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml pretrained_models/simple_res50_256x192.pth ./data/1.mp4
++ CONFIG=configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml
++ CKPT=pretrained_models/simple_res50_256x192.pth
++ VIDEO=./data/1.mp4
++ OUTDIR=./examples/res
++ python scripts/demo_inference.py --cfg configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml --checkpoint pretrained_models/simple_res50_256x192.pth --video ./data/1.mp4 --outdir ./examples/res --detector yolo --save_img --save_video
Traceback (most recent call last):
  File "scripts/demo_inference.py", line 13, in <module>
    from alphapose.models import builder
  File "/home/tianxin/AlphaPose/alphapose/models/__init__.py", line 1, in <module>
    from .fastpose import FastPose
  File "/home/tianxin/AlphaPose/alphapose/models/fastpose.py", line 10, in <module>
    from .layers.SE_Resnet import SEResnet
  File "/home/tianxin/AlphaPose/alphapose/models/layers/SE_Resnet.py", line 9, in <module>
    from .dcn import DeformConv, ModulatedDeformConv
  File "/home/tianxin/AlphaPose/alphapose/models/layers/dcn/__init__.py", line 1, in <module>
    from .deform_conv import (DeformConv, DeformConvPack, ModulatedDeformConv,
  File "/home/tianxin/AlphaPose/alphapose/models/layers/dcn/deform_conv.py", line 9, in <module>
    from . import deform_conv_cuda
ImportError: /home/tianxin/AlphaPose/alphapose/models/layers/dcn/deform_conv_cuda.cpython-36m-x86_64-linux-gnu.so: undefined symbol: _ZN2at7getTypeERKNS_6TensorE

Current environment：
pytorch==1.2.0
torchvision==0.4.0
cudatoolkit=9.2

[Fang-Haoshu]

Hi, I think it should work with this cuda version. It's pretty weird. Maybe you can google related errors.

[Fang-Haoshu]

Would it be related to GCC version? What is the GCC version. You may want to check this link:
facebookresearch/maskrcnn-benchmark#25

[GuoHaiYang123]

I have already tried, and this error will still appear on the gcc5.4which version<6.0 @Fang-Haoshu @starry-xin

[starry-xin]

The GCC version is 4.9.3 now, and I will check and try to update it. Thanks!

[IMBINGO95]

I met this problem too！ But I dont kown how to solve it.
Current environment：
pytorch==1.1.0
torchvision==0.3.0
cudatoolkit=10.0
gcc==4.9.3
@Fang-Haoshu

[Fang-Haoshu]

@IMBINGO95 How about using GCC5.4?

[IMBINGO95]

@IMBINGO95 How about using GCC5.4?

I tried on another machine with the same environment except gcc==5.4 ,but it failed again !
I don't konw why .

[IMBINGO95]

Is there a requirement for the CUDA version?
ImportError happens when I run the demo：

$ ./scripts/inference.sh configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml pretrained_models/simple_res50_256x192.pth ./data/1.mp4
++ CONFIG=configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml
++ CKPT=pretrained_models/simple_res50_256x192.pth
++ VIDEO=./data/1.mp4
++ OUTDIR=./examples/res
++ python scripts/demo_inference.py --cfg configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml --checkpoint pretrained_models/simple_res50_256x192.pth --video ./data/1.mp4 --outdir ./examples/res --detector yolo --save_img --save_video
Traceback (most recent call last):
  File "scripts/demo_inference.py", line 13, in <module>
    from alphapose.models import builder
  File "/home/tianxin/AlphaPose/alphapose/models/__init__.py", line 1, in <module>
    from .fastpose import FastPose
  File "/home/tianxin/AlphaPose/alphapose/models/fastpose.py", line 10, in <module>
    from .layers.SE_Resnet import SEResnet
  File "/home/tianxin/AlphaPose/alphapose/models/layers/SE_Resnet.py", line 9, in <module>
    from .dcn import DeformConv, ModulatedDeformConv
  File "/home/tianxin/AlphaPose/alphapose/models/layers/dcn/__init__.py", line 1, in <module>
    from .deform_conv import (DeformConv, DeformConvPack, ModulatedDeformConv,
  File "/home/tianxin/AlphaPose/alphapose/models/layers/dcn/deform_conv.py", line 9, in <module>
    from . import deform_conv_cuda
ImportError: /home/tianxin/AlphaPose/alphapose/models/layers/dcn/deform_conv_cuda.cpython-36m-x86_64-linux-gnu.so: undefined symbol: _ZN2at7getTypeERKNS_6TensorE

Current environment：
pytorch==1.2.0
torchvision==0.4.0
cudatoolkit=9.2

Have you solved the probem? If you did, can you tell me how to solve it ？ Please， thank you very much!

[starry-xin]

solve

I change the environment into pytorch==1.1.0, torchvision==0.3.0, cudatoolkit=10.0 and install cuda==10.0 instead of cuda==9.2 which have existed in system. Then the error disappeared.
And my GCC version is 4.9.3, which seems feasible. @IMBINGO95

[starry-xin]

The GCC version is 4.9.3 now, and I will check and try to update it. Thanks!

It dosen't seem to matter.

[Fang-Haoshu]

I see, thanks! @starry-xin

[RSKothari]

So is the solution to move to CUDA 10?

[starry-xin]

So is the solution to move to CUDA 10?

I guess so in the environment required.