Add temporary fix for CUDA issue in Dockerfile (NVIDIA#5421)

Signed-off-by: Yu Yao <yuya@nvidia.com> Signed-off-by: Yu Yao <yuya@nvidia.com>
1-800-BAD-CODE · Nov 15, 2022 · 0718b17 · 0718b17
1 parent 1b5fac4
commit 0718b17
Showing 1 changed file with 4 additions and 0 deletions.
diff --git a/Dockerfile b/Dockerfile
@@ -100,5 +100,9 @@ COPY tests /workspace/nemo/tests
 COPY tutorials /workspace/nemo/tutorials
 # COPY README.rst LICENSE /workspace/nemo/
 
+# Temporary fix CUDA issue
+RUN sed -i "s/, all_gpu_ids//g" /opt/conda/lib/python3.8/site-packages/pytorch_lightning/accelerators/cuda.py
+RUN sed -i "s/all_gpu_ids =/\# all_gpu_ids =/g" /opt/conda/lib/python3.8/site-packages/pytorch_lightning/accelerators/cuda.py
+
 RUN printf "#!/bin/bash\njupyter lab --no-browser --allow-root --ip=0.0.0.0" >> start-jupyter.sh && \
     chmod +x start-jupyter.sh