diff --git a/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/benchmark.txt b/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/benchmark.txt index 4fd7ff9960daad4d165b24e7652341b0653a4bf2..b7d56042588e7df75a05c7a9976781acdae8ae2a 100644 --- a/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/benchmark.txt +++ b/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/benchmark.txt @@ -20,6 +20,9 @@ apptainer exec --nv -B ${WORK_DIR}/data:/data -B ${WORK_DIR}/results:/results -- apptainer exec --nv -B ${WORK_DIR}/data:/data -B ${WORK_DIR}/results:/results --pwd /workspace/nnunet_pyt $CONTAINER_DIR python scripts/benchmark.py --mode predict --gpus 1 --dim 2 --batch_size 256 --amp + + + ################# Issues ################# # Known issue https://github.com/NVIDIA/DeepLearningExamples/issues/1113 ImportError: cannot import name 'get_num_classes' from 'torchmetrics.utilities.data' (/opt/conda/lib/python3.8/site-packages/torchmetrics/utilities/data.py) diff --git a/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/generate_benchmark_jobs.sh b/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/generate_benchmark_jobs.sh index d2175c1a7528c91fd8d168bae83e82b1a39a641d..b4b874e55d43e77b845bb5ca7015849408d464c1 100644 --- a/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/generate_benchmark_jobs.sh +++ b/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/generate_benchmark_jobs.sh @@ -23,7 +23,7 @@ EOT if [ "${6}" = "thin" ]; then cat <<EOT >> $SBATCH_DIR #SBATCH -C "thin" -##SBATCH --reservation=devel +#SBATCH --reservation=nodeimage EOT else cat <<EOT >> $SBATCH_DIR diff --git a/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/submit_benchmark_jobs.sh b/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/submit_benchmark_jobs.sh index a28efd6816c872589cb8e91d199b7d6d454d2fce..e5968719ae7996c3ff886c5f85811a40145cb68e 100644 --- a/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/submit_benchmark_jobs.sh +++ b/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/submit_benchmark_jobs.sh @@ -1,13 +1,13 @@ #!/bin/bash set -e -WORK_DIR=/proj/nsc_testing/xuan/berzelius-benchmarks/PyTorch/Segmentation/nnUNet +WORK_DIR=/proj/nsc_testing/xuan/berzelius-benchmarks/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet benchmark_modes=("train" "predict") node_types=("thin" "fat") dim=2 for nodes in {1..1}; do - for gpus in {1,8}; do + for gpus in {1..8}; do for benchmark_mode in "${benchmark_modes[@]}"; do for node_type in "${node_types[@]}"; do