aws · marfriaz · Jan 17, 2025
diff --git a/advanced_functionality/causal-inference/causal-inference-container.ipynb b/advanced_functionality/causal-inference/causal-inference-container.ipynb
@@ -183,7 +183,7 @@
     "\n",
     "When training large models with huge amounts of data, you'll typically use big data tools, like Amazon Athena, AWS Glue, or Amazon EMR, to create your data in S3. For the purposes of this example, we are using a heart failure dataset of 299 patients. \n",
     "\n",
-    "Davide Chicco, Giuseppe Jurman: \u201cMachine learning can predict survival of patients with heart failure from serum creatinine and ejection fraction alone\u201d. BMC Medical Informatics and Decision Making 20, 16 (2020).\n",
+    "Davide Chicco, Giuseppe Jurman: “Machine learning can predict survival of patients with heart failure from serum creatinine and ejection fraction alone”. BMC Medical Informatics and Decision Making 20, 16 (2020).\n",
     "[Web Link](\n",
     "https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-020-1023-5)\n",
     "\n",
@@ -382,13 +382,21 @@
     "account = sess.boto_session.client(\"sts\").get_caller_identity()[\"Account\"]\n",
     "region = sess.boto_session.region_name\n",
     "image = \"{}.dkr.ecr.{}.amazonaws.com/causal-nex-container:latest\".format(account, region)\n",
+    "default_bucket = sess.default_bucket()\n",
+    "default_bucket_prefix = sess.default_bucket_prefix\n",
+    "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    s3_output_path = f\"s3://{default_bucket}/{default_bucket_prefix}/output\"\n",
+    "else:\n",
+    "    s3_output_path = f\"s3://{default_bucket}/output\"\n",
     "\n",
     "bn = sage.estimator.Estimator(\n",
     "    image_uri=image,\n",
     "    role=role,\n",
     "    instance_count=1,\n",
     "    instance_type=\"ml.c4.2xlarge\",\n",
-    "    output_path=\"s3://{}/output\".format(sess.default_bucket()),\n",
+    "    output_path=s3_output_path,\n",
     "    sagemaker_session=sess,\n",
     ")\n",
     "\n",
@@ -502,7 +510,7 @@
     "- Does treatment X help to cure the disease?\n",
     "- What happens if we change the type Y?\n",
     "\n",
-    "Actually doing the intervention might be unfeasible or unethical \u2014 side-stepping actual interventions and still getting at causal effects is the whole point of this approach to [causal inference](https://fabiandablander.com/r/Causal-Inference.html).\n",
+    "Actually doing the intervention might be unfeasible or unethical — side-stepping actual interventions and still getting at causal effects is the whole point of this approach to [causal inference](https://fabiandablander.com/r/Causal-Inference.html).\n",
     "To read more about interventions [go here](https://medium.data4sci.com/causal-inference-part-ix-interventions-c3f94190191d) or [here](https://www.cmu.edu/dietrich/philosophy/docs/scheines/PSA2006.pdf). For the deeper understanding we recommend reading \"Book of Why\" by Judea Pearl.\n",
     "\n",
     "Some example questions that can be answered with Causal Analysis are:\n",
@@ -641,4 +649,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 4
-}
+}
diff --git a/async-inference/Async-Inference-Walkthrough-SageMaker-Python-SDK.ipynb b/async-inference/Async-Inference-Walkthrough-SageMaker-Python-SDK.ipynb
@@ -218,6 +218,12 @@
    "outputs": [],
    "source": [
     "bucket_prefix = \"DEMO-async-inference\"\n",
+    "default_bucket_prefix = sm_session.default_bucket_prefix\n",
+    "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    bucket_prefix = f\"{default_bucket_prefix}/{bucket_prefix}\"\n",
+    "\n",
     "resource_name = \"AsyncInferenceDemo-{}-{}\""
    ]
   },
@@ -418,7 +424,7 @@
     "        # additional instances before the effects of previous activities are visible.\n",
     "        # You can configure the length of time based on your instance startup time or other application needs.\n",
     "        # ScaleInCooldown - The amount of time, in seconds, after a scale in activity completes before another scale in activity can start.\n",
-    "        \"ScaleOutCooldown\": 300  # ScaleOutCooldown - The amount of time, in seconds, after a scale out activity completes before another scale out activity can start.\n",
+    "        \"ScaleOutCooldown\": 300,  # ScaleOutCooldown - The amount of time, in seconds, after a scale out activity completes before another scale out activity can start.\n",
     "        # 'DisableScaleIn': True|False - ndicates whether scale in by the target tracking policy is disabled.\n",
     "        # If the value is true , scale in is disabled and the target tracking policy won't remove capacity from the scalable resource.\n",
     "    },\n",

diff --git a/async-inference/Async-Inference-Walkthrough.ipynb b/async-inference/Async-Inference-Walkthrough.ipynb
@@ -220,6 +220,12 @@
    "outputs": [],
    "source": [
     "bucket_prefix = \"async-inference-demo\"\n",
+    "default_bucket_prefix = sm_session.default_bucket_prefix\n",
+    "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    bucket_prefix = f\"{default_bucket_prefix}/{bucket_prefix}\"\n",
+    "\n",
     "resource_name = \"AsyncInferenceDemo-{}-{}\""
    ]
   },
@@ -448,7 +454,7 @@
     "        # additional instances before the effects of previous activities are visible.\n",
     "        # You can configure the length of time based on your instance startup time or other application needs.\n",
     "        # ScaleInCooldown - The amount of time, in seconds, after a scale in activity completes before another scale in activity can start.\n",
-    "        \"ScaleOutCooldown\": 300  # ScaleOutCooldown - The amount of time, in seconds, after a scale out activity completes before another scale out activity can start.\n",
+    "        \"ScaleOutCooldown\": 300,  # ScaleOutCooldown - The amount of time, in seconds, after a scale out activity completes before another scale out activity can start.\n",
     "        # 'DisableScaleIn': True|False - ndicates whether scale in by the target tracking policy is disabled.\n",
     "        # If the value is true , scale in is disabled and the target tracking policy won't remove capacity from the scalable resource.\n",
     "    },\n",

diff --git a/async-inference/HuggingFace-Async-Inference-Walkthrough.ipynb b/async-inference/HuggingFace-Async-Inference-Walkthrough.ipynb
@@ -141,6 +141,7 @@
     "sagemaker_client = boto_session.client(\"sagemaker\")\n",
     "sm_runtime = boto_session.client(\"sagemaker-runtime\")\n",
     "s3_bucket = sm_session.default_bucket()\n",
+    "default_bucket_prefix = sm_session.default_bucket_prefix\n",
     "current_timestamp = strftime(\"%m-%d-%H-%M\", gmtime())\n",
     "logger.info(f\"Region = {region}\")\n",
     "logger.info(f\"Role = {role}\")"
@@ -289,7 +290,15 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "endpoint_config_name = model_name"
+    "endpoint_config_name = model_name\n",
+    "\n",
+    "input_s3_location = f\"s3://{s3_bucket}/input/input.json\"\n",
+    "output_s3_location = f\"s3://{s3_bucket}/output\"\n",
+    "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    input_s3_location = f\"s3://{s3_bucket}/{default_bucket_prefix}/input/input.json\"\n",
+    "    output_s3_location = f\"s3://{s3_bucket}/{default_bucket_prefix}/output\""
    ]
   },
   {
@@ -324,7 +333,7 @@
     "    ],\n",
     "    AsyncInferenceConfig={\n",
     "        \"OutputConfig\": {\n",
-    "            \"S3OutputPath\": f\"s3://{s3_bucket}/output\",\n",
+    "            \"S3OutputPath\": output_s3_location,\n",
     "            # Optionally specify Amazon SNS topics\n",
     "            # \"NotificationConfig\": {\n",
     "            #   \"SuccessTopic\": \"arn:aws:sns:us-east-2:123456789012:MyTopic\",\n",
@@ -390,7 +399,6 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "input_s3_location = f\"s3://{s3_bucket}/input/input.json\"\n",
     "print(input_s3_location)"
    ]
   },
@@ -555,4 +563,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 4
-}
+}
diff --git a/async-inference/Transcription_on_SM_endpoint.ipynb b/async-inference/Transcription_on_SM_endpoint.ipynb
@@ -191,8 +191,11 @@
     "sess = sagemaker.Session()\n",
     "\n",
     "sagemaker_session_bucket = None\n",
+    "default_bucket_prefix = None\n",
+    "\n",
     "if sagemaker_session_bucket is None and sess is not None:\n",
     "    sagemaker_session_bucket = sess.default_bucket()\n",
+    "    default_bucket_prefix = sess.default_bucket_prefix\n",
     "\n",
     "try:\n",
     "    role = sagemaker.get_execution_role()\n",
@@ -214,7 +217,13 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "s3_location = f\"s3://{sagemaker_session_bucket}/whisper/model/model.tar.gz\""
+    "s3_location = f\"s3://{sagemaker_session_bucket}/whisper/model/model.tar.gz\"\n",
+    "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    s3_location = (\n",
+    "        f\"s3://{sagemaker_session_bucket}/{default_bucket_prefix}/whisper/model/model.tar.gz\"\n",
+    "    )"
    ]
   },
   {
@@ -332,6 +341,12 @@
     "\n",
     "async_endpoint_name = name_from_base(\"whisper-large-v2-custom-asyc\")\n",
     "\n",
+    "s3_output_path = \"async_inference/output\"\n",
+    "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    s3_output_path = f\"{default_bucket_prefix}/async_inference/output\"\n",
+    "\n",
     "# create Hugging Face Model Class\n",
     "huggingface_model = HuggingFaceModel(\n",
     "    model_data=s3_location,  # path to your model and script\n",
@@ -344,7 +359,7 @@
     "# create async endpoint configuration\n",
     "async_config = AsyncInferenceConfig(\n",
     "    output_path=s3_path_join(\n",
-    "        \"s3://\", sagemaker_session_bucket, \"async_inference/output\"\n",
+    "        \"s3://\", sagemaker_session_bucket, s3_output_path\n",
     "    ),  # Where our results will be stored\n",
     "    # Add nofitication SNS if needed\n",
     "    notification_config={\n",
@@ -461,7 +476,7 @@
     "            \"Statistic\": \"Average\",\n",
     "        },\n",
     "        \"ScaleInCooldown\": 300,  # ScaleInCooldown - The amount of time, in seconds, after a scale-in activity completes before another scale in activity can start.\n",
-    "        \"ScaleOutCooldown\": 300  # ScaleOutCooldown - The amount of time, in seconds, after a scale-out activity completes before another scale out activity can start.\n",
+    "        \"ScaleOutCooldown\": 300,  # ScaleOutCooldown - The amount of time, in seconds, after a scale-out activity completes before another scale out activity can start.\n",
     "        # 'DisableScaleIn': True|False - indicates whether scale in by the target tracking policy is disabled.\n",
     "        # If the value is true, scale-in is disabled and the target tracking policy won't remove capacity from the scalable resource.\n",
     "    },\n",

diff --git a/.../cv/realtime/Triton/multi-model/tensorflow-backend/triton-cv-mme-tensorflow-backend.ipynb b/.../cv/realtime/Triton/multi-model/tensorflow-backend/triton-cv-mme-tensorflow-backend.ipynb
@@ -181,6 +181,7 @@
     "role = sagemaker.get_execution_role()\n",
     "bucket_name = sm_session.default_bucket()\n",
     "region = boto3.Session().region_name\n",
+    "default_bucket_prefix = sm_session.default_bucket_prefix\n",
     "\n",
     "print(f\"Default IAM Role: {role}\")\n",
     "print(f\"Default S3 Bucket: {bucket_name}\")\n",
@@ -308,6 +309,10 @@
    "source": [
     "prefix = \"triton-mme\"\n",
     "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    prefix = f\"{default_bucket_prefix}/{prefix}\"\n",
+    "\n",
     "! mkdir -p models/$prefix/MNIST1/1\n",
     "! cp models/SavedModel/00000000 --recursive ./models/$prefix/MNIST1/1/model.savedmodel/"
    ]

diff --git a/.../realtime/Triton/single-model/resnet_onnx-backend/resnet_onnx_backend_SME_triton_v2.ipynb b/.../realtime/Triton/single-model/resnet_onnx-backend/resnet_onnx_backend_SME_triton_v2.ipynb
@@ -113,8 +113,13 @@
     "runtime_sm_client = boto3.client(\"sagemaker-runtime\")\n",
     "sagemaker_session = sagemaker.Session(boto_session=boto3.Session())\n",
     "bucket = sagemaker_session.default_bucket()\n",
+    "default_bucket_prefix = sagemaker_session.default_bucket_prefix\n",
     "prefix = \"resnet50-sme-gpu-onnx\"\n",
     "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    prefix = f\"{default_bucket_prefix}/{prefix}\"\n",
+    "\n",
     "# endpoint variables\n",
     "sm_model_name = f\"{prefix}-mdl-{ts}\"\n",
     "endpoint_config_name = f\"{prefix}-epc-{ts}\"\n",

diff --git a/...cv/realtime/Triton/single-model/tensorflow-backend/triton-cv-sme-tensorflow-backend.ipynb b/...cv/realtime/Triton/single-model/tensorflow-backend/triton-cv-sme-tensorflow-backend.ipynb
@@ -146,6 +146,7 @@
     "sm_session = sagemaker.Session()\n",
     "role = sagemaker.get_execution_role()\n",
     "bucket_name = sm_session.default_bucket()\n",
+    "default_bucket_prefix = sm_session.default_bucket_prefix\n",
     "region = boto3.Session().region_name\n",
     "\n",
     "print(f\"Default IAM Role: {role}\")\n",
@@ -304,6 +305,10 @@
    "source": [
     "prefix = \"triton-sme\"\n",
     "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    prefix = f\"{default_bucket_prefix}/{prefix}\"\n",
+    "\n",
     "# Reorganize the model structure that is required by Triton Server.\n",
     "! mkdir -p model/$prefix/MNIST/1\n",
     "! cp model/SavedModel/00000000 --recursive ./model/$prefix/MNIST/1/model.savedmodel/"

diff --git a/.../triton/multi-model/resnet_pytorch_python-backend/resnet_pytorch_python_backend_MME.ipynb b/.../triton/multi-model/resnet_pytorch_python-backend/resnet_pytorch_python_backend_MME.ipynb
@@ -147,8 +147,14 @@
     "runtime_sm_client = boto3.client(\"sagemaker-runtime\")\n",
     "sagemaker_session = sagemaker.Session(boto_session=boto3.Session())\n",
     "bucket = sagemaker_session.default_bucket()\n",
+    "default_bucket_prefix = sagemaker_session.default_bucket_prefix\n",
+    "\n",
     "prefix = \"resnet50-mme-gpu\"\n",
     "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    prefix = f\"{default_bucket_prefix}/{prefix}\"\n",
+    "\n",
     "# endpoint variables\n",
     "sm_model_name = f\"{prefix}-mdl-{ts}\"\n",
     "endpoint_config_name = f\"{prefix}-epc-{ts}\"\n",

diff --git a/...esnet_onnx_pytorch_tensorRT-backend/resnet_onnx_pytorch_tensorRT_backend_MME_triton.ipynb b/...esnet_onnx_pytorch_tensorRT-backend/resnet_onnx_pytorch_tensorRT_backend_MME_triton.ipynb
@@ -139,8 +139,13 @@
     "runtime_sm_client = boto3.client(\"sagemaker-runtime\")\n",
     "sagemaker_session = sagemaker.Session(boto_session=boto3.Session())\n",
     "bucket = sagemaker_session.default_bucket()\n",
+    "default_bucket_prefix = sagemaker_session.default_bucket_prefix\n",
     "prefix = \"resnet50-mme-gpu-onnx-tensorRT-pyTorch\"\n",
     "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    prefix = f\"{default_bucket_prefix}/{prefix}\"\n",
+    "\n",
     "# endpoint variables\n",
     "sm_model_name = f\"{prefix}-mdl-{ts}\"\n",
     "endpoint_config_name = f\"{prefix}-epc-{ts}\"\n",

diff --git a/inference/generativeai/deepspeed/GPT-J-6B_DJLServing_with_PySDK.ipynb b/inference/generativeai/deepspeed/GPT-J-6B_DJLServing_with_PySDK.ipynb
@@ -100,6 +100,11 @@
     "\n",
     "# bucket = session.default_bucket()      # bucket to house artifacts\n",
     "# s3_location = f\"s3://{bucket}/djl-serving/gpt-j-6B\"\n",
+    "# default_bucket_prefix = session.default_bucket_prefix\n",
+    "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "# if default_bucket_prefix:\n",
+    "#     s3_location = f\"s3://{bucket}/{default_bucket_prefix}/djl-serving/gpt-j-6B\"\n",
     "# S3Uploader.upload(\"gpt-j-6B\", s3_location)"
    ]
   },

diff --git a/inference/generativeai/huggingface-multimodal/stability-cascade/DeployStableCascade.ipynb b/inference/generativeai/huggingface-multimodal/stability-cascade/DeployStableCascade.ipynb
@@ -325,9 +325,18 @@
    "outputs": [],
    "source": [
     "# upload model.tar.gz to s3\n",
+    "default_bucket = sess.default_bucket()\n",
+    "default_bucket_prefix = sess.default_bucket_prefix\n",
+    "\n",
+    "s3_upload_path = f\"s3://{default_bucket}/stable-cascade\"\n",
+    "\n",
+    "# If a default bucket prefix is specified, append it to the s3 path\n",
+    "if default_bucket_prefix:\n",
+    "    s3_upload_path = f\"s3://{default_bucket}/{default_bucket_prefix}/stable-cascade\"\n",
+    "\n",
     "s3_model_uri = S3Uploader.upload(\n",
     "    local_path=\"model.tar.gz\",\n",
-    "    desired_s3_uri=f\"s3://{sess.default_bucket()}/stable-cascade\",\n",
+    "    desired_s3_uri=s3_upload_path,\n",
     ")\n",
     "print(f\"model uploaded to: {s3_model_uri}\")"
    ]