Fix smddp notebook account number (#3431)

* add updates to pass account id * notebook changes * some clean ups Co-authored-by: atqy <[email protected]>
aws · Jun 7, 2022 · 4e181db · 4e181db
1 parent 1927b11
commit 4e181db
Show file tree

Hide file tree

Showing 7 changed files with 61 additions and 40 deletions.
diff --git a/training/distributed_training/pytorch/data_parallel/bert/DeepLearningExamples b/training/distributed_training/pytorch/data_parallel/bert/DeepLearningExamples
diff --git a/training/distributed_training/pytorch/data_parallel/bert/Dockerfile b/training/distributed_training/pytorch/data_parallel/bert/Dockerfile
@@ -1,6 +1,7 @@
+ARG dlc_account_id
 ARG region
 
-FROM 763104351884.dkr.ecr.${region}.amazonaws.com/pytorch-training:1.10.2-gpu-py38-cu113-ubuntu20.04-sagemaker
+FROM ${dlc_account_id}.dkr.ecr.${region}.amazonaws.com/pytorch-training:1.10.2-gpu-py38-cu113-ubuntu20.04-sagemaker
 
 ARG WORK_DIR="apex_build"
 RUN pip --no-cache-dir --no-cache install h5py boto3 'git+https://github.com/NVIDIA/dllogger' tqdm requests; \

diff --git a/training/distributed_training/pytorch/data_parallel/bert/build_and_push.sh b/training/distributed_training/pytorch/data_parallel/bert/build_and_push.sh
@@ -7,39 +7,44 @@
 
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
 
-if [ "$#" -eq 3 ]; then
-    region=$1
-    image=$2
-    tag=$3
+if [ "$#" -eq 4 ]; then
+    dlc_account_id=$1
+    region=$2
+    image=$3
+    tag=$4
 else
-    echo "usage: $0 <aws-region> $1 <image-repo> $2 <image-tag>"
+    echo "usage: $0 <dlc-account-id> $1 <aws-region> $2 <image-repo> $3 <image-tag>"
     exit 1
 fi
+
 # Get the account number associated with the current IAM credentials
 account=$(aws sts get-caller-identity --query Account --output text)
+
 if [ $? -ne 0 ]
 then
     exit 255
 fi
 
+
 fullname="${account}.dkr.ecr.${region}.amazonaws.com/${image}:${tag}"
+
 # If the repository doesn't exist in ECR, create it.
 aws ecr describe-repositories --region ${region} --repository-names "${image}" > /dev/null 2>&1
 if [ $? -ne 0 ]; then
+    echo "creating ECR repository : ${fullname} "
     aws ecr create-repository --region ${region} --repository-name "${image}" > /dev/null
 fi
-# Build the docker image locally with the image name and then push it to ECR
-# with the full name.
-# login ECR for the current account
 
-aws ecr get-login-password --region ${region} | docker login --username AWS --password-stdin ${account}.dkr.ecr.${region}.amazonaws.com
+aws ecr get-login-password --region ${region} \
+| docker login --username AWS --password-stdin ${account}.dkr.ecr.${region}.amazonaws.com
 
-docker build ${DIR}/ -t ${image} -f ${DIR}/Dockerfile  --build-arg region=${region}
+docker build . -t ${image} -f ${DIR}/Dockerfile  --build-arg dlc_account_id=${dlc_account_id} --build-arg region=${region}
 docker tag ${image} ${fullname}
 docker push ${fullname}
+
 if [ $? -eq 0 ]; then
 	echo "Amazon ECR URI: ${fullname}"
 else
 	echo "Error: Image build and push failed"
 	exit 1
-fi
+fi
diff --git a/...ng/distributed_training/pytorch/data_parallel/bert/pytorch_smdataparallel_bert_demo.ipynb b/...ng/distributed_training/pytorch/data_parallel/bert/pytorch_smdataparallel_bert_demo.ipynb
@@ -134,8 +134,17 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "%%time\n",
-    "! chmod +x build_and_push.sh; bash build_and_push.sh {region} {image} {tag}"
+    "dlc_account_id = 763104351884  # By default, set the account ID used for most regions"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "! aws ecr get-login-password --region {region} | docker login --username AWS --password-stdin {dlc_account_id}.dkr.ecr.{region}.amazonaws.com\n",
+    "! chmod +x build_and_push.sh; bash build_and_push.sh {dlc_account_id} {region} {image} {tag}"
    ]
   },
   {
@@ -316,9 +325,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "conda_pytorch_p36",
+   "display_name": "conda_amazonei_pytorch_latest_p36",
    "language": "python",
-   "name": "conda_pytorch_p36"
+   "name": "conda_amazonei_pytorch_latest_p36"
   },
   "language_info": {
    "codemirror_mode": {
@@ -330,7 +339,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.10"
+   "version": "3.6.13"
   }
  },
  "nbformat": 4,

diff --git a/training/distributed_training/pytorch/data_parallel/maskrcnn/Dockerfile b/training/distributed_training/pytorch/data_parallel/maskrcnn/Dockerfile
@@ -1,6 +1,7 @@
+ARG dlc_account_id
 ARG region
 
-FROM 763104351884.dkr.ecr.${region}.amazonaws.com/pytorch-training:1.10.2-gpu-py38-cu113-ubuntu20.04-sagemaker
+FROM ${dlc_account_id}.dkr.ecr.${region}.amazonaws.com/pytorch-training:1.10.2-gpu-py38-cu113-ubuntu20.04-sagemaker
 
 RUN pip install ninja yacs cython matplotlib tqdm opencv-python pybind11==2.5.0 'git+https://github.com/NVIDIA/dllogger'
 

diff --git a/training/distributed_training/pytorch/data_parallel/maskrcnn/build_and_push.sh b/training/distributed_training/pytorch/data_parallel/maskrcnn/build_and_push.sh
@@ -7,12 +7,13 @@
 
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
 
-if [ "$#" -eq 3 ]; then
-    region=$1
-    image=$2
-    tag=$3
+if [ "$#" -eq 4 ]; then
+    dlc_account_id=$1
+    region=$2
+    image=$3
+    tag=$4
 else
-    echo "usage: $0 <aws-region> $1 <image-repo> $2 <image-tag>"
+    echo "usage: $0 <dlc-account-id> $1 <aws-region> $2 <image-repo> $3 <image-tag>"
     exit 1
 fi
 
@@ -34,16 +35,16 @@ if [ $? -ne 0 ]; then
     aws ecr create-repository --region ${region} --repository-name "${image}" > /dev/null
 fi
 
-$(aws ecr get-login --no-include-email --region ${region}  --registry-ids 763104351884)
-docker build ${DIR}/ -t ${image} -f ${DIR}/Dockerfile  --build-arg region=${region}
-docker tag ${image} ${fullname}
+aws ecr get-login-password --region ${region} \
+| docker login --username AWS --password-stdin ${account}.dkr.ecr.${region}.amazonaws.com
 
-# Get the login command from ECR and execute it directly
-$(aws ecr get-login --region ${region} --no-include-email)
+docker build . -t ${image} -f ${DIR}/Dockerfile  --build-arg dlc_account_id=${dlc_account_id} --build-arg region=${region}
+docker tag ${image} ${fullname}
 docker push ${fullname}
+
 if [ $? -eq 0 ]; then
 	echo "Amazon ECR URI: ${fullname}"
 else
 	echo "Error: Image build and push failed"
 	exit 1
-fi
+fi
diff --git a/...ibuted_training/pytorch/data_parallel/maskrcnn/pytorch_smdataparallel_maskrcnn_demo.ipynb b/...ibuted_training/pytorch/data_parallel/maskrcnn/pytorch_smdataparallel_maskrcnn_demo.ipynb
@@ -136,8 +136,17 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "%%time\n",
-    "! chmod +x build_and_push.sh; bash build_and_push.sh {region} {image} {tag}"
+    "dlc_account_id = 763104351884  # By default, set the account ID used for most regions"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "! aws ecr get-login-password --region {region} | docker login --username AWS --password-stdin {dlc_account_id}.dkr.ecr.{region}.amazonaws.com\n",
+    "! chmod +x build_and_push.sh; bash build_and_push.sh {dlc_account_id} {region} {image} {tag}"
    ]
   },
   {
@@ -299,13 +308,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3.7.4 64-bit ('base': conda)",
-   "metadata": {
-    "interpreter": {
-     "hash": "dca0ade3e726a953b501b15e8e990130d2b7799f14cfd9f4271676035ebe5511"
-    }
-   },
-   "name": "python3"
+   "display_name": "conda_amazonei_pytorch_latest_p36",
+   "language": "python",
+   "name": "conda_amazonei_pytorch_latest_p36"
   },
   "language_info": {
    "codemirror_mode": {
@@ -317,7 +322,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.4"
+   "version": "3.6.13"
   }
  },
  "nbformat": 4,