experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Explicitly specify CUDA and CUDNN versions (#15328)

Browse Source 
A workaround to the base problem causing https://github.com/tensorflow/tensorflow/issues/15291
This commit is contained in:
Austin Anderson 2017-12-12 21:24:11 -08:00 committed by Gunhan Gulsoy
parent e2174cc943
commit d7dbf21bda
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
tensorflow/tools/docker/Dockerfile.devel-gpu
View File

@ -81,7 +81,8 @@ ENV CI_BUILD_PYTHON python
ENV LD_LIBRARY_PATH /usr/local/cuda/extras/CUPTI/lib64:$LD_LIBRARY_PATH
ENV TF_NEED_CUDA 1
ENV TF_CUDA_COMPUTE_CAPABILITIES=3.0,3.5,5.2,6.0,6.1
ENV TF_CUDA_VERSION=9.0
ENV TF_CUDNN_VERSION=7
RUN ln -s /usr/local/cuda/lib64/stubs/libcuda.so /usr/local/cuda/lib64/stubs/libcuda.so.1 && \
LD_LIBRARY_PATH=/usr/local/cuda/lib64/stubs:${LD_LIBRARY_PATH} \