Update 3 files

- /PyTorch/Segmentation/nnUNet/run_benchmark_single_node.sh
- /PyTorch/Segmentation/nnUNet/build.txt
- /PyTorch/Segmentation/nnUNet/nnunet_for_pytorch_21.11.0.def

PyTorch/Segmentation/nnUNet/build.txt

@@ -32,5 +32,8 @@ apptainer exec --nv --pwd /workspace/nnunet_pyt $CONTAINER_DIR python scripts/be
 apptainer exec --nv --pwd /workspace/nnunet_pyt $CONTAINER_DIR python scripts/benchmark.py --mode predict --gpus 1 --dim 2 --batch_size 256 --amp
 
 
+apptainer exec --nv -B ${WORK_DIR}/data:/data -B ${WORK_DIR}/results:/results $CONTAINER_DIR bash -c "cd /workspace/nnunet_pyt && python scripts/benchmark.py --mode train --gpus 1  --dim 2 --batch_size 256 --amp" 
+
+
 # Known issue 
 https://github.com/NVIDIA/DeepLearningExamples/issues/1113

PyTorch/Segmentation/nnUNet/nnunet_for_pytorch_21.11.0.def

@@ -16,11 +16,12 @@ cd /workspace/nnunet_pyt
 wget --content-disposition https://api.ngc.nvidia.com/v2/resources/nvidia/nnunet_for_pytorch/versions/$VERSION/zip -O nnunet_for_pytorch_$VERSION.zip
 unzip nnunet_for_pytorch_21.11.0.zip
 
+sed 's/pytorch-lightning==1.3.8/pytorch-lightning==1.4.0/g' requirements.txt > temp_file && mv temp_file requirements.txt
 pip install --disable-pip-version-check -r requirements.txt
 pip install monai==0.8.0 --no-dependencies
-pip uninstall -y torchtext torchmetrics
+pip uninstall -y torchtext 
 # Fix for https://github.com/NVIDIA/DeepLearningExamples/issues/1113
-pip install torchmetrics==0.6.0
+#pip install torchmetrics==0.6.0
 
 curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip"
 unzip -qq awscliv2.zip

PyTorch/Segmentation/nnUNet/run_benchmark_single_node.sh

@@ -3,6 +3,7 @@
 MODULE_NAME=nnunet_for_pytorch
 MODULE_VERSION=21.11.0
 WORK_DIR=/proj/nsc_testing/xuan/berzelius-benchmarks/PyTorch/Segmentation/nnUNet
+CONTAINER_DIR=/proj/nsc_testing/xuan/containers/${MODULE_NAME}_${MODULE_VERSION}.sif
 SBATCH_DIR=$WORK_DIR/sbatch_scripts/benchmark_dim${1}_nodes${2}_gpus${3}_batchsize${4}_iteration${5}.sbatch
 SBATCH_OUT_DIR=$WORK_DIR/sbatch_out/benchmark_dim${1}_nodes${2}_gpus${3}_batchsize${4}_iteration${5}.out
 LOG_DIR=benchmark_dim${1}_nodes${2}_gpus${3}_batchsize${4}_amp_iteration${5}.json