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

nvcc fatal : Value 'sm_20' is not defined for option 'gpu-architecture' #1580

Closed
awhan opened this issue Nov 15, 2017 · 3 comments
Closed

nvcc fatal : Value 'sm_20' is not defined for option 'gpu-architecture' #1580

awhan opened this issue Nov 15, 2017 · 3 comments
Labels
gpu Using spaCy on GPU install Installation issues 🔮 thinc spaCy's machine learning library Thinc

Comments

@awhan
Copy link

awhan commented Nov 15, 2017

Spacy fails to install on archlinux distro due to error in installation of thinc.

I used the following to install pip install --user spacy. Error that I could glean from the log file is

nvcc fatal : Value 'sm_20' is not defined for option 'gpu-architecture'

while the following message is what I noted in the terminal

Command "/usr/bin/python -u -c "import setuptools, tokenize;file='/tmp/pip-build-mqkbeylk/thinc/setup.py';f=getattr(tokenize, 'open', open)(file);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, file, 'exec'))" install --record /tmp/pip-s1ppwpnp-record/install-record.txt --single-version-externally-managed --compile --user --prefix=" failed with error code 1 in /tmp/pip-build-mqkbeylk/thinc/

output of nvcc --version is

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2017 NVIDIA Corporation
Built on Fri_Sep__1_21:08:03_CDT_2017
Cuda compilation tools, release 9.0, V9.0.176

pip install log is here

Your Environment

  • Operating System: uname -a is Linux zero 4.13.12-1-ARCH failed on build #1 SMP PREEMPT Wed Nov 8 11:54:06 CET 2017 x86_64 GNU/Linux
  • Python Version Used: Python 3.6.3
  • spaCy Version Used: could not install (attempted to install 2.0.2)
  • Environment Information: don't know how to generate without spacy installed
@ines ines added the install Installation issues label Nov 15, 2017
@honnibal
Copy link
Member

honnibal commented Nov 15, 2017

Fixing this in Thinc, thanks. In the meantime you can set the environment variable CUDA9=1 to tell Thinc to drop the offending compiler flag.

@honnibal
Copy link
Member

Should now be fixed in Thinc.

@ines ines added the 🔮 thinc spaCy's machine learning library Thinc label Nov 15, 2017
@ines ines closed this as completed Nov 15, 2017
@ines ines added the gpu Using spaCy on GPU label Dec 22, 2017
@lock
Copy link

lock bot commented May 8, 2018

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.

@lock lock bot locked as resolved and limited conversation to collaborators May 8, 2018
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
gpu Using spaCy on GPU install Installation issues 🔮 thinc spaCy's machine learning library Thinc
Projects
None yet
Development

No branches or pull requests

3 participants