diff --git a/NVIDIA/DeepLearningExamples/PyTorch/README.md b/NVIDIA/DeepLearningExamples/PyTorch/README.md
index 06f65902c4d5e8d60cdfe574e7763c40fd3369b9..798bc662205958dde14d42851b47b70040ae331d 100644
--- a/NVIDIA/DeepLearningExamples/PyTorch/README.md
+++ b/NVIDIA/DeepLearningExamples/PyTorch/README.md
@@ -42,7 +42,7 @@ bash submit_benchmark_jobs.sh
 
 ### Known issues
 
-#### Isssue 1
+#### Isssue 1 (21.11.0)
 https://github.com/NVIDIA/DeepLearningExamples/issues/1113
 
 When running the container, an error occurred:
@@ -74,6 +74,15 @@ OSError: [Errno 22] Invalid argument
 
 We need to comment out the L32-33 in the `main.py` to fix it.
 
-#### Issue 2 
+#### Issue 2 (21.11.0)
 
 Muiti-node jobs is not supported yet in 21.11.0 but only in the most recent code on GitHub.
+
+
+#### Issue 3 (latest)
+
+The last line of the `Dockerfile` has to change to:
+
+```
+cp utils/instance_norm.py /opt/conda/lib/python3.8/site-packages/apex/normalization
+```
\ No newline at end of file
diff --git a/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/nnunet_for_pytorch_latest.def b/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/nnunet_for_pytorch_latest.def
index 96321096f112377615e3dafc4cb6ebba9ea5a1e7..5545cd4f24621eac1ee09f2eb4d419ed2bf941b9 100644
--- a/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/nnunet_for_pytorch_latest.def
+++ b/NVIDIA/DeepLearningExamples/PyTorch/Segmentation/nnUNet/nnunet_for_pytorch_latest.def
@@ -26,4 +26,4 @@ unzip -qq awscliv2.zip
 ./aws/install
 rm -rf awscliv2.zip aws
 
-cp utils/instance_norm.py /usr/local/lib/python3.8/dist-packages/apex/normalization
+cp utils/instance_norm.py /opt/conda/lib/python3.8/site-packages/apex/normalization