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

error: class "at::Type" has no member "scalarType" #267

Closed
DinghaoZhang opened this issue Apr 23, 2019 · 6 comments · Fixed by #272
Closed

error: class "at::Type" has no member "scalarType" #267

DinghaoZhang opened this issue Apr 23, 2019 · 6 comments · Fixed by #272

Comments

@DinghaoZhang
Copy link

I can't setup this, error is:
csrc/type_shim.h(13): error: class "at::Type" has no member "scalarType"
1 error detected in the compilation of "/tmp/tmpxft_00004318_00000000-6_multi_tensor_scale_kernel.cpp1.ii".
error: command '/usr/local/cuda/bin/nvcc' failed with exit status 1
who know how to fix it?

@ptrblck
Copy link
Collaborator

ptrblck commented Apr 23, 2019

Which PyTorch version are you using?
You can check it with python -c "import torch; print(torch.__version__)"

In case your PyTorch version is older than 0.4, you should update it using the install instructions given on the PyTorch website.

@Stonesjtu
Copy link

Stonesjtu commented Apr 23, 2019

The master branch of pytorch now does break the backward compatibility of cpp-extension due to a breaking change, you should check out to release version.

@ptrblck
Copy link
Collaborator

ptrblck commented Apr 23, 2019

@Stonesjtu Could you post a link to the issue if possible?

@Stonesjtu
Copy link

Stonesjtu commented Apr 23, 2019

@ptrblck Sure, Check this out : pytorch/pytorch#17991. My original text does not support should be wrong, it's just an API change, sorry for that.

@ptrblck
Copy link
Collaborator

ptrblck commented Apr 23, 2019

Thanks for letting us know! :)
I'll take a look into this issue so that we can build again from PyTorch master.

@mcarilli
Copy link
Contributor

mcarilli commented Apr 24, 2019

@ptrblck's fix is in flight (#272), should be merged in the next couple days. We also need to ensure backward compatibility with versions of Pytorch from 1.0 to present, and get rid of some deprecation warnings with current master.

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

Successfully merging a pull request may close this issue.

4 participants