Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Hard problem! #6

Open
jemmryx opened this issue Nov 4, 2019 · 1 comment
Open

Hard problem! #6

jemmryx opened this issue Nov 4, 2019 · 1 comment

Comments

@jemmryx
Copy link

jemmryx commented Nov 4, 2019

Why this error happening?
I got this error but run this code in another GPU secucessfully. The environments are the same????

File "/home/jiaxin/maxout_race/paragraph_race_4.06/model.py", line 335, in forward
out, _ = scatter_max(energy, ext_src_seq, out=out)
File "/home/jiaxin/maxout_race/env3.5/lib/python3.5/site-packages/torch_scatter/max.py", line 109, in scatter_max
return ScatterMax.apply(out, src, index, dim)
File "/home/jiaxin/maxout_race/env3.5/lib/python3.5/site-packages/torch_scatter/max.py", line 13, in forward
func(src, index, out, arg, dim)
RuntimeError: index.type().is_cuda() INTERNAL ASSERT FAILED at cuda/scatter.cpp:35, please report a bug to PyTorch. index must be CUDA tensor (scatter_max at cuda/scatter.cpp:35)
frame #0: c10::Error::Error(c10::SourceLocation, std::string const&) + 0x33 (0x7f3aab35e813 in /home/jiaxin/maxout_race/env3.5/lib/python3.5/site-packages/torch/lib/libc10.so)
frame #1: scatter_max(at::Tensor, at::Tensor, at::Tensor, at::Tensor, long) + 0x2b7 (0x7f3a90e994b7 in /home/jiaxin/maxout_race/env3.5/lib/python3.5/site-packages/torch_scatter/scatter_cuda.cpython-35m-x86_64-linux-gnu.so)
frame #2: + 0x28748 (0x7f3a90ea6748 in /home/jiaxin/maxout_race/env3.5/lib/python3.5/site-packages/torch_scatter/scatter_cuda.cpython-35m-x86_64-linux-gnu.so)
frame #3: + 0x288ee (0x7f3a90ea68ee in /home/jiaxin/maxout_race/env3.5/lib/python3.5/site-packages/torch_scatter/scatter_cuda.cpython-35m-x86_64-linux-gnu.so)
frame #4: + 0x24bac (0x7f3a90ea2bac in /home/jiaxin/maxout_race/env3.5/lib/python3.5/site-packages/torch_scatter/scatter_cuda.cpython-35m-x86_64-linux-gnu.so)

frame #8: python() [0x4e3423]
frame #11: THPFunction_apply(_object*, _object*) + 0xa4f (0x7f3aac3f04af in /home/jiaxin/maxout_race/env3.5/lib/python3.5/site-packages/torch/lib/libtorch_python.so)
frame #14: python() [0x539a13]
frame #17: python() [0x4e3537]
frame #21: python() [0x4e3423]
frame #23: python() [0x4f08be]
frame #25: python() [0x55fbf6]
frame #31: python() [0x539a13]
frame #33: python() [0x6292c2]
frame #38: __libc_start_main + 0xf0 (0x7f3ab0539830 in /lib/x86_64-linux-gnu/libc.so.6)

@seanie12
Copy link
Owner

Did you install pytorch-scatter and run the test?

It seems like the error is related to pytorch-scatter.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants