NCv3 VM retirement update (#3696)

* ncv3 retirement update

* review comment

Author: PratibhaShrivastav18

cli/endpoints/batch/deploy-models/imagenet-classifier/deploy-and-run.sh

@@ -26,7 +26,7 @@ az ml model create --name $MODEL_NAME --path "model"
 
 echo "Creating compute with GPU"
 # <create_compute>
-az ml compute create -n gpu-cluster --type amlcompute --size STANDARD_NC6s_v3 --min-instances 0 --max-instances 2
+az ml compute create -n gpu-cluster --type amlcompute --size STANDARD_NC4AS_T4_V3 --min-instances 0 --max-instances 2
 # </create_compute>
 
 echo "Creating batch endpoint $ENDPOINT_NAME"

cli/endpoints/online/triton/single-model/create-managed-deployment.yaml

@@ -7,4 +7,4 @@ model:
   path: ./models
   type: triton_model
 instance_count: 1
-instance_type: Standard_NC6s_v3
+instance_type: STANDARD_NC4AS_T4_V3

cli/foundation-models/system/finetune/image-classification/multiclass-classification/hftransformers-fridgeobjects-multiclass-classification.sh

@@ -13,8 +13,8 @@ compute_cluster_finetune="sample-finetune-cluster-gpu"
 compute_model_evaluation="sample-finetune-cluster-gpu"
 # If above compute cluster does not exist, create it with the following vm size
 compute_model_import_sku="Standard_D12"
-compute_finetune_sku="Standard_NC6s_v3"
-compute_model_evaluation_sku="Standard_NC6s_v3"
+compute_finetune_sku="STANDARD_NC4AS_T4_V3"
+compute_model_evaluation_sku="STANDARD_NC4AS_T4_V3"
 
 # This is the number of GPUs in a single node of the selected 'vm_size' compute. 
 # Setting this to less than the number of GPUs will result in underutilized GPUs, taking longer to train.

cli/foundation-models/system/finetune/image-classification/multilabel-classification/hftransformers-fridgeobjects-multilabel-classification.sh

@@ -13,8 +13,8 @@ compute_cluster_finetune="sample-finetune-cluster-gpu"
 compute_model_evaluation="sample-finetune-cluster-gpu"
 # If above compute cluster does not exist, create it with the following vm size
 compute_model_import_sku="Standard_D12"
-compute_finetune_sku="Standard_NC6s_v3"
-compute_model_evaluation_sku="Standard_NC6s_v3"
+compute_finetune_sku="STANDARD_NC4AS_T4_V3"
+compute_model_evaluation_sku="STANDARD_NC4AS_T4_V3"
 
 # This is the number of GPUs in a single node of the selected 'vm_size' compute. 
 # Setting this to less than the number of GPUs will result in underutilized GPUs, taking longer to train.

cli/foundation-models/system/finetune/image-instance-segmentation/mmdetection-fridgeobjects-instance-segmentation.sh

@@ -15,8 +15,8 @@ compute_cluster_finetune="sample-finetune-cluster-gpu"
 compute_model_evaluation="sample-finetune-cluster-gpu"
 # If above compute cluster does not exist, create it with the following vm size
 compute_model_import_sku="Standard_D12"
-compute_finetune_sku="Standard_NC6s_v3"
-compute_model_evaluation_sku="Standard_NC6s_v3"
+compute_finetune_sku="STANDARD_NC4AS_T4_V3"
+compute_model_evaluation_sku="STANDARD_NC4AS_T4_V3"
 
 # This is the number of GPUs in a single node of the selected 'vm_size' compute. 
 # Setting this to less than the number of GPUs will result in underutilized GPUs, taking longer to train.

cli/foundation-models/system/finetune/image-object-detection/mmdetection-fridgeobjects-detection.sh

@@ -14,8 +14,8 @@ compute_cluster_finetune="sample-finetune-cluster-gpu"
 compute_model_evaluation="sample-finetune-cluster-gpu"
 # If above compute cluster does not exist, create it with the following vm size
 compute_model_import_sku="Standard_D12"
-compute_finetune_sku="Standard_NC6s_v3"
-compute_model_evaluation_sku="Standard_NC6s_v3"
+compute_finetune_sku="STANDARD_NC4AS_T4_V3"
+compute_model_evaluation_sku="STANDARD_NC4AS_T4_V3"
 
 # This is the number of GPUs in a single node of the selected 'vm_size' compute. 
 # Setting this to less than the number of GPUs will result in underutilized GPUs, taking longer to train.

cli/foundation-models/system/finetune/question-answering/extractive-qa.sh

@@ -10,7 +10,7 @@ registry_name="azureml"
 
 compute_cluster="gpu-cluster-big"
 # if above compute cluster does not exist, create it with the following vm size
-compute_sku="Standard_NC24rs_v3"
+compute_sku="STANDARD_NC4AS_T4_V3"
 # This is the number of GPUs in a single node of the selected 'vm_size' compute. 
 # Setting this to less than the number of GPUs will result in underutilized GPUs, taking longer to train.
 # Setting this to more than the number of GPUs will result in an error.

cli/foundation-models/system/finetune/summarization/news-summary.sh

@@ -10,7 +10,7 @@ registry_name="azureml"
 
 compute_cluster="gpu-cluster-big"
 # if above compute cluster does not exist, create it with the following vm size
-compute_sku="Standard_NC24rs_v3"
+compute_sku="STANDARD_NC4AS_T4_V3"
 # This is the number of GPUs in a single node of the selected 'vm_size' compute. 
 # Setting this to less than the number of GPUs will result in underutilized GPUs, taking longer to train.
 # Setting this to more than the number of GPUs will result in an error.

cli/foundation-models/system/finetune/text-classification/emotion-detection.sh

@@ -10,7 +10,7 @@ registry_name="azureml"
 
 compute_cluster="gpu-cluster-big"
 # if above compute cluster does not exist, create it with the following vm size
-compute_sku="Standard_NC24rs_v3"
+compute_sku="STANDARD_NC4AS_T4_V3"
 # This is the number of GPUs in a single node of the selected 'vm_size' compute. 
 # Setting this to less than the number of GPUs will result in underutilized GPUs, taking longer to train.
 # Setting this to more than the number of GPUs will result in an error.

cli/foundation-models/system/finetune/text-to-image/deploy.yaml

@@ -1,5 +1,5 @@
 $schema: https://azuremlschemas.azureedge.net/latest/managedOnlineDeployment.schema.json
-instance_type: Standard_NC6s_v3
+instance_type: STANDARD_NC4AS_T4_V3
 instance_count: 1
 liveness_probe:
   initial_delay: 180