sagemaker-hyperpod 3.0.1__tar.gz → 3.0.2__tar.gz

{sagemaker_hyperpod-3.0.1/src/sagemaker_hyperpod.egg-info → sagemaker_hyperpod-3.0.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sagemaker-hyperpod
-Version: 3.0.1
+Version: 3.0.2
 Summary: Amazon SageMaker HyperPod SDK and CLI
 Home-page: https://github.com/aws/sagemaker-hyperpod-cli
 Author: Amazon Web Services
@@ -212,8 +212,8 @@ hyp create hyp-pytorch-job \
     --version 1.0 \
     --job-name test-pytorch-job \
     --image pytorch/pytorch:latest \
-    --command '["python", "train.py"]' \
-    --args '["--epochs", "10", "--batch-size", "32"]' \
+    --command '[python, train.py]' \
+    --args '[--epochs=10, --batch-size=32]' \
     --environment '{"PYTORCH_CUDA_ALLOC_CONF": "max_split_size_mb:32"}' \
     --pull-policy "IfNotPresent" \
     --instance-type ml.p4d.24xlarge \
@@ -224,9 +224,8 @@ hyp create hyp-pytorch-job \
     --queue-name "training-queue" \
     --priority "high" \
     --max-retry 3 \
-    --volumes '["data-vol", "model-vol", "checkpoint-vol"]' \
-    --persistent-volume-claims '["shared-data-pvc", "model-registry-pvc"]' \
-    --output-s3-uri s3://my-bucket/model-artifacts
+    --volume name=model-data,type=hostPath,mount_path=/data,path=/data \
+    --volume name=training-output,type=pvc,mount_path=/data,claim_name=my-pvc,read_only=false
 ```
 
 Key required parameters explained:
@@ -235,8 +234,6 @@ Key required parameters explained:
 
     --image: Docker image containing your training environment
 
-This command starts a training job named test-pytorch-job. The --output-s3-uri specifies where the trained model artifacts will be stored, for example, s3://my-bucket/model-artifacts. Note this location, as you’ll need it for deploying the custom model.
-
 ### Inference 
 
 #### Creating a JumpstartModel Endpoint
@@ -311,9 +308,10 @@ Along with the CLI, we also have SDKs available that can perform the training an
 
 ```
 
-from sagemaker.hyperpod import HyperPodPytorchJob
-from sagemaker.hyperpod.job 
-import ReplicaSpec, Template, Spec, Container, Resources, RunPolicy, Metadata
+from sagemaker.hyperpod.training import HyperPodPytorchJob
+from sagemaker.hyperpod.training
+import ReplicaSpec, Template, Spec, Containers, Resources, RunPolicy
+from sagemaker.hyperpod.common.config import Metadata
 
 # Define job specifications
 nproc_per_node = "1"  # Number of processes per node
@@ -328,7 +326,7 @@ replica_specs =
             (
                 containers =
                 [
-                    Container
+                    Containers
                     (
                         # Container name
                         name="container-name",  
@@ -369,8 +367,6 @@ pytorch_job = HyperPodPytorchJob
     replica_specs = replica_specs,     
     # Run policy
     run_policy = run_policy,           
-    # S3 location for artifacts
-    output_s3_uri="s3://my-bucket/model-artifacts"  
 )
 # Launch the job
 pytorch_job.create()  
@@ -390,24 +386,21 @@ Pre-trained Jumpstart models can be gotten from https://sagemaker.readthedocs.io
 from sagemaker.hyperpod.inference.config.hp_jumpstart_endpoint_config import Model, Server, SageMakerEndpoint, TlsConfig
 from sagemaker.hyperpod.inference.hp_jumpstart_endpoint import HPJumpStartEndpoint
 
-model = Model(
-    model_id="deepseek-llm-r1-distill-qwen-1-5b",
-    model_version="2.0.4"
+model=Model(
+    model_id='deepseek-llm-r1-distill-qwen-1-5b',
+    model_version='2.0.4',
 )
-
-server = Server(
-    instance_type="ml.g5.8xlarge"
+server=Server(
+    instance_type='ml.g5.8xlarge',
 )
+endpoint_name=SageMakerEndpoint(name='<my-endpoint-name>')
+tls_config=TlsConfig(tls_certificate_output_s3_uri='s3://<my-tls-bucket>')
 
-endpoint_name = SageMakerEndpoint(name="endpoint-jumpstart")
-
-tls_config = TlsConfig(tls_certificate_output_s3_uri="s3://sample-bucket")
-
-js_endpoint = HPJumpStartEndpoint(
+js_endpoint=HPJumpStartEndpoint(
     model=model,
     server=server,
     sage_maker_endpoint=endpoint_name,
-    tls_config=tls_config
+    tls_config=tls_config,
 )
 
 js_endpoint.create()
@@ -423,51 +416,51 @@ print(response)
 ```
 
 
-#### Creating a Custom Inference Endpoint 
+#### Creating a Custom Inference Endpoint (with S3)
 
 ```
-from sagemaker.hyperpod.inference.config.hp_custom_endpoint_config import Model, Server, SageMakerEndpoint, TlsConfig, EnvironmentVariables
-from sagemaker.hyperpod.inference.hp_custom_endpoint import HPCustomEndpoint
+from sagemaker.hyperpod.inference.config.hp_endpoint_config import CloudWatchTrigger, Dimensions, AutoScalingSpec, Metrics, S3Storage, ModelSourceConfig, TlsConfig, EnvironmentVariables, ModelInvocationPort, ModelVolumeMount, Resources, Worker
+from sagemaker.hyperpod.inference.hp_endpoint import HPEndpoint
 
-model = Model(
-    model_source_type="s3",
-    model_location="test-pytorch-job/model.tar.gz",
-    s3_bucket_name="my-bucket",
-    s3_region="us-east-2",
-    prefetch_enabled=True
+model_source_config = ModelSourceConfig(
+    model_source_type='s3',
+    model_location="<my-model-folder-in-s3>",
+    s3_storage=S3Storage(
+        bucket_name='<my-model-artifacts-bucket>',
+        region='us-east-2',
+    ),
 )
 
-server = Server(
-    instance_type="ml.g5.8xlarge",
-    image_uri="763104351884.dkr.ecr.us-east-2.amazonaws.com/huggingface-pytorch-tgi-inference:2.4.0-tgi2.3.1-gpu-py311-cu124-ubuntu22.04-v2.0",
-    container_port=8080,
-    model_volume_mount_name="model-weights"
-)
+environment_variables = [
+    EnvironmentVariables(name="HF_MODEL_ID", value="/opt/ml/model"),
+    EnvironmentVariables(name="SAGEMAKER_PROGRAM", value="inference.py"),
+    EnvironmentVariables(name="SAGEMAKER_SUBMIT_DIRECTORY", value="/opt/ml/model/code"),
+    EnvironmentVariables(name="MODEL_CACHE_ROOT", value="/opt/ml/model"),
+    EnvironmentVariables(name="SAGEMAKER_ENV", value="1"),
+]
 
-resources = {
-    "requests": {"cpu": "30000m", "nvidia.com/gpu": 1, "memory": "100Gi"},
-    "limits": {"nvidia.com/gpu": 1}
-}
-
-env = EnvironmentVariables(
-    HF_MODEL_ID="/opt/ml/model",
-    SAGEMAKER_PROGRAM="inference.py",
-    SAGEMAKER_SUBMIT_DIRECTORY="/opt/ml/model/code",
-    MODEL_CACHE_ROOT="/opt/ml/model",
-    SAGEMAKER_ENV="1"
+worker = Worker(
+    image='763104351884.dkr.ecr.us-east-2.amazonaws.com/huggingface-pytorch-tgi-inference:2.4.0-tgi2.3.1-gpu-py311-cu124-ubuntu22.04-v2.0',
+    model_volume_mount=ModelVolumeMount(
+        name='model-weights',
+    ),
+    model_invocation_port=ModelInvocationPort(container_port=8080),
+    resources=Resources(
+            requests={"cpu": "30000m", "nvidia.com/gpu": 1, "memory": "100Gi"},
+            limits={"nvidia.com/gpu": 1}
+    ),
+    environment_variables=environment_variables,
 )
 
-endpoint_name = SageMakerEndpoint(name="endpoint-custom-pytorch")
-
-tls_config = TlsConfig(tls_certificate_output_s3_uri="s3://sample-bucket")
+tls_config=TlsConfig(tls_certificate_output_s3_uri='s3://<my-tls-bucket-name>')
 
-custom_endpoint = HPCustomEndpoint(
-    model=model,
-    server=server,
-    resources=resources,
-    environment=env,
-    sage_maker_endpoint=endpoint_name,
+custom_endpoint = HPEndpoint(
+    endpoint_name='<my-endpoint-name>',
+    instance_type='ml.g5.8xlarge',
+    model_name='deepseek15b-test-model-name',  
     tls_config=tls_config,
+    model_source_config=model_source_config,
+    worker=worker,
 )
 
 custom_endpoint.create()
@@ -484,19 +477,17 @@ print(response)
 #### Managing an Endpoint 
 
 ```
-endpoint_iterator = HPJumpStartEndpoint.list()
-for endpoint in endpoint_iterator:
-    print(endpoint.name, endpoint.status)
+endpoint_list = HPEndpoint.list()
+print(endpoint_list[0])
 
-logs = js_endpoint.get_logs()
-print(logs)
+print(custom_endpoint.get_operator_logs(since_hours=0.5))
 
 ```
 
 #### Deleting an Endpoint 
 
 ```
-js_endpoint.delete()
+custom_endpoint.delete()
 
 ```
 

{sagemaker_hyperpod-3.0.1 → sagemaker_hyperpod-3.0.2}/README.md

@@ -158,8 +158,8 @@ hyp create hyp-pytorch-job \
     --version 1.0 \
     --job-name test-pytorch-job \
     --image pytorch/pytorch:latest \
-    --command '["python", "train.py"]' \
-    --args '["--epochs", "10", "--batch-size", "32"]' \
+    --command '[python, train.py]' \
+    --args '[--epochs=10, --batch-size=32]' \
     --environment '{"PYTORCH_CUDA_ALLOC_CONF": "max_split_size_mb:32"}' \
     --pull-policy "IfNotPresent" \
     --instance-type ml.p4d.24xlarge \
@@ -170,9 +170,8 @@ hyp create hyp-pytorch-job \
     --queue-name "training-queue" \
     --priority "high" \
     --max-retry 3 \
-    --volumes '["data-vol", "model-vol", "checkpoint-vol"]' \
-    --persistent-volume-claims '["shared-data-pvc", "model-registry-pvc"]' \
-    --output-s3-uri s3://my-bucket/model-artifacts
+    --volume name=model-data,type=hostPath,mount_path=/data,path=/data \
+    --volume name=training-output,type=pvc,mount_path=/data,claim_name=my-pvc,read_only=false
 ```
 
 Key required parameters explained:
@@ -181,8 +180,6 @@ Key required parameters explained:
 
     --image: Docker image containing your training environment
 
-This command starts a training job named test-pytorch-job. The --output-s3-uri specifies where the trained model artifacts will be stored, for example, s3://my-bucket/model-artifacts. Note this location, as you’ll need it for deploying the custom model.
-
 ### Inference 
 
 #### Creating a JumpstartModel Endpoint
@@ -257,9 +254,10 @@ Along with the CLI, we also have SDKs available that can perform the training an
 
 ```
 
-from sagemaker.hyperpod import HyperPodPytorchJob
-from sagemaker.hyperpod.job 
-import ReplicaSpec, Template, Spec, Container, Resources, RunPolicy, Metadata
+from sagemaker.hyperpod.training import HyperPodPytorchJob
+from sagemaker.hyperpod.training
+import ReplicaSpec, Template, Spec, Containers, Resources, RunPolicy
+from sagemaker.hyperpod.common.config import Metadata
 
 # Define job specifications
 nproc_per_node = "1"  # Number of processes per node
@@ -274,7 +272,7 @@ replica_specs =
             (
                 containers =
                 [
-                    Container
+                    Containers
                     (
                         # Container name
                         name="container-name",  
@@ -315,8 +313,6 @@ pytorch_job = HyperPodPytorchJob
     replica_specs = replica_specs,     
     # Run policy
     run_policy = run_policy,           
-    # S3 location for artifacts
-    output_s3_uri="s3://my-bucket/model-artifacts"  
 )
 # Launch the job
 pytorch_job.create()  
@@ -336,24 +332,21 @@ Pre-trained Jumpstart models can be gotten from https://sagemaker.readthedocs.io
 from sagemaker.hyperpod.inference.config.hp_jumpstart_endpoint_config import Model, Server, SageMakerEndpoint, TlsConfig
 from sagemaker.hyperpod.inference.hp_jumpstart_endpoint import HPJumpStartEndpoint
 
-model = Model(
-    model_id="deepseek-llm-r1-distill-qwen-1-5b",
-    model_version="2.0.4"
+model=Model(
+    model_id='deepseek-llm-r1-distill-qwen-1-5b',
+    model_version='2.0.4',
 )
-
-server = Server(
-    instance_type="ml.g5.8xlarge"
+server=Server(
+    instance_type='ml.g5.8xlarge',
 )
+endpoint_name=SageMakerEndpoint(name='<my-endpoint-name>')
+tls_config=TlsConfig(tls_certificate_output_s3_uri='s3://<my-tls-bucket>')
 
-endpoint_name = SageMakerEndpoint(name="endpoint-jumpstart")
-
-tls_config = TlsConfig(tls_certificate_output_s3_uri="s3://sample-bucket")
-
-js_endpoint = HPJumpStartEndpoint(
+js_endpoint=HPJumpStartEndpoint(
     model=model,
     server=server,
     sage_maker_endpoint=endpoint_name,
-    tls_config=tls_config
+    tls_config=tls_config,
 )
 
 js_endpoint.create()
@@ -369,51 +362,51 @@ print(response)
 ```
 
 
-#### Creating a Custom Inference Endpoint 
+#### Creating a Custom Inference Endpoint (with S3)
 
 ```
-from sagemaker.hyperpod.inference.config.hp_custom_endpoint_config import Model, Server, SageMakerEndpoint, TlsConfig, EnvironmentVariables
-from sagemaker.hyperpod.inference.hp_custom_endpoint import HPCustomEndpoint
+from sagemaker.hyperpod.inference.config.hp_endpoint_config import CloudWatchTrigger, Dimensions, AutoScalingSpec, Metrics, S3Storage, ModelSourceConfig, TlsConfig, EnvironmentVariables, ModelInvocationPort, ModelVolumeMount, Resources, Worker
+from sagemaker.hyperpod.inference.hp_endpoint import HPEndpoint
 
-model = Model(
-    model_source_type="s3",
-    model_location="test-pytorch-job/model.tar.gz",
-    s3_bucket_name="my-bucket",
-    s3_region="us-east-2",
-    prefetch_enabled=True
+model_source_config = ModelSourceConfig(
+    model_source_type='s3',
+    model_location="<my-model-folder-in-s3>",
+    s3_storage=S3Storage(
+        bucket_name='<my-model-artifacts-bucket>',
+        region='us-east-2',
+    ),
 )
 
-server = Server(
-    instance_type="ml.g5.8xlarge",
-    image_uri="763104351884.dkr.ecr.us-east-2.amazonaws.com/huggingface-pytorch-tgi-inference:2.4.0-tgi2.3.1-gpu-py311-cu124-ubuntu22.04-v2.0",
-    container_port=8080,
-    model_volume_mount_name="model-weights"
-)
+environment_variables = [
+    EnvironmentVariables(name="HF_MODEL_ID", value="/opt/ml/model"),
+    EnvironmentVariables(name="SAGEMAKER_PROGRAM", value="inference.py"),
+    EnvironmentVariables(name="SAGEMAKER_SUBMIT_DIRECTORY", value="/opt/ml/model/code"),
+    EnvironmentVariables(name="MODEL_CACHE_ROOT", value="/opt/ml/model"),
+    EnvironmentVariables(name="SAGEMAKER_ENV", value="1"),
+]
 
-resources = {
-    "requests": {"cpu": "30000m", "nvidia.com/gpu": 1, "memory": "100Gi"},
-    "limits": {"nvidia.com/gpu": 1}
-}
-
-env = EnvironmentVariables(
-    HF_MODEL_ID="/opt/ml/model",
-    SAGEMAKER_PROGRAM="inference.py",
-    SAGEMAKER_SUBMIT_DIRECTORY="/opt/ml/model/code",
-    MODEL_CACHE_ROOT="/opt/ml/model",
-    SAGEMAKER_ENV="1"
+worker = Worker(
+    image='763104351884.dkr.ecr.us-east-2.amazonaws.com/huggingface-pytorch-tgi-inference:2.4.0-tgi2.3.1-gpu-py311-cu124-ubuntu22.04-v2.0',
+    model_volume_mount=ModelVolumeMount(
+        name='model-weights',
+    ),
+    model_invocation_port=ModelInvocationPort(container_port=8080),
+    resources=Resources(
+            requests={"cpu": "30000m", "nvidia.com/gpu": 1, "memory": "100Gi"},
+            limits={"nvidia.com/gpu": 1}
+    ),
+    environment_variables=environment_variables,
 )
 
-endpoint_name = SageMakerEndpoint(name="endpoint-custom-pytorch")
-
-tls_config = TlsConfig(tls_certificate_output_s3_uri="s3://sample-bucket")
+tls_config=TlsConfig(tls_certificate_output_s3_uri='s3://<my-tls-bucket-name>')
 
-custom_endpoint = HPCustomEndpoint(
-    model=model,
-    server=server,
-    resources=resources,
-    environment=env,
-    sage_maker_endpoint=endpoint_name,
+custom_endpoint = HPEndpoint(
+    endpoint_name='<my-endpoint-name>',
+    instance_type='ml.g5.8xlarge',
+    model_name='deepseek15b-test-model-name',  
     tls_config=tls_config,
+    model_source_config=model_source_config,
+    worker=worker,
 )
 
 custom_endpoint.create()
@@ -430,19 +423,17 @@ print(response)
 #### Managing an Endpoint 
 
 ```
-endpoint_iterator = HPJumpStartEndpoint.list()
-for endpoint in endpoint_iterator:
-    print(endpoint.name, endpoint.status)
+endpoint_list = HPEndpoint.list()
+print(endpoint_list[0])
 
-logs = js_endpoint.get_logs()
-print(logs)
+print(custom_endpoint.get_operator_logs(since_hours=0.5))
 
 ```
 
 #### Deleting an Endpoint 
 
 ```
-js_endpoint.delete()
+custom_endpoint.delete()
 
 ```
 

{sagemaker_hyperpod-3.0.1 → sagemaker_hyperpod-3.0.2}/pyproject.toml

@@ -5,7 +5,7 @@ build-backend = "setuptools.build_meta"
 [project]
 dynamic = ["dependencies"]
 name = "sagemaker-hyperpod"
-version = "3.0.1"
+version = "3.0.2"
 description = "Amazon SageMaker HyperPod SDK and CLI"
 readme = "README.md"
 requires-python = ">=3.8"

{sagemaker_hyperpod-3.0.1 → sagemaker_hyperpod-3.0.2}/setup.py

@@ -47,7 +47,7 @@ for root, dirs, files in os.walk(
 setup(
     data_files=sagemaker_hyperpod_recipes,
     name="sagemaker-hyperpod",
-    version="3.0.1",
+    version="3.0.2",
     description="Amazon SageMaker HyperPod SDK and CLI",
     long_description=open("README.md").read(),
     long_description_content_type="text/markdown",

{sagemaker_hyperpod-3.0.1 → sagemaker_hyperpod-3.0.2}/src/sagemaker/hyperpod/cli/commands/cluster.py

@@ -120,7 +120,7 @@ def list_cluster(
     debug: bool,
     namespace: Optional[List],
 ):
-    """List SageMaker Hyperpod Clusters with cluster metadata.
+    """List SageMaker Hyperpod Clusters with metadata.
 
     Example Usage:
     1. List clusters with JSON output: hyperpod get-clusters -n hyperpod-ns-test-team
@@ -233,7 +233,7 @@ def list_cluster(
         print(tabulate(cluster_capacities, headers=headers, tablefmt="presto"))
     elif output == OutputFormat.JSON.value:
         json_list = [dict(zip(headers, value)) for value in cluster_capacities]
-        _restructure_output(json_list, namespace)
+        json_list = _restructure_output(json_list, namespace)
         print(json.dumps(json_list, indent=4))
 
 
@@ -379,23 +379,34 @@ def _get_hyperpod_clusters(sm_client: boto3.client) -> List[str]:
 
 
 def _restructure_output(summary_list, namespaces):
-    if not namespaces:
-        return
+    cluster_dict = dict()
 
     for node_summary in summary_list:
-        node_summary["Namespaces"] = {}
-        for ns in namespaces:
-            available_accelerators = node_summary[
-                ns + AVAILABLE_ACCELERATOR_DEVICES_KEY
-            ]
-            total_accelerators = node_summary[ns + TOTAL_ACCELERATOR_DEVICES_KEY]
-            quota_accelerator_info = {
-                AVAILABLE_ACCELERATOR_DEVICES_KEY: available_accelerators,
-                TOTAL_ACCELERATOR_DEVICES_KEY: total_accelerators,
+        cluster_name = node_summary["Cluster"]
+        if cluster_name not in cluster_dict:
+            cluster_dict[cluster_name] = {
+                "Cluster": cluster_name,
+                "Instances": []
             }
-            node_summary["Namespaces"][ns] = quota_accelerator_info
-            node_summary.pop(ns + AVAILABLE_ACCELERATOR_DEVICES_KEY, None)
-            node_summary.pop(ns + TOTAL_ACCELERATOR_DEVICES_KEY, None)
+        node_summary.pop("Cluster")
+        if namespaces:
+            node_summary["Namespaces"] = {}
+            for ns in namespaces:
+                available_accelerators = node_summary[
+                    ns + AVAILABLE_ACCELERATOR_DEVICES_KEY
+                ]
+                total_accelerators = node_summary[ns + TOTAL_ACCELERATOR_DEVICES_KEY]
+                quota_accelerator_info = {
+                    AVAILABLE_ACCELERATOR_DEVICES_KEY: available_accelerators,
+                    TOTAL_ACCELERATOR_DEVICES_KEY: total_accelerators,
+                }
+                node_summary["Namespaces"][ns] = quota_accelerator_info
+                node_summary.pop(ns + AVAILABLE_ACCELERATOR_DEVICES_KEY, None)
+                node_summary.pop(ns + TOTAL_ACCELERATOR_DEVICES_KEY, None)
+        cluster_dict[cluster_name]["Instances"].append(node_summary)
+
+    return list(cluster_dict.values())
+
 
 
 def _aggregate_nodes_info(
@@ -553,7 +564,7 @@ def get_cluster_context(
     debug: bool,
 ) -> Tuple[Any, str]:
     """
-    Get all the context related to the current set Cluster
+    Get context related to the current set cluster.
 
     Args:
         debug (bool): Enable debug mode.
@@ -584,7 +595,7 @@ def get_cluster_context(
 @click.option("--prometheus", is_flag=True, help="Returns Prometheus Workspace URL")
 @click.option("--list", is_flag=True, help="Returns list of available metrics")
 def get_monitoring(grafana: bool, prometheus: bool, list: bool) -> None:
-    """Get monitoring configurations for Hyperpod cluster"""
+    """Get monitoring configurations for Hyperpod cluster."""
     try:
         if not any([grafana, prometheus, list]):
             print("Error: Please select at least one option")