edit aws_marketplace/using_model_packages/generic_sample_notebook/A_generic_sample_notebook_to_perform_inference_on_ML_model_packages_from_AWS_Marketplace.ipynb

Author: EC2 Default User

aws_marketplace/using_model_packages/generic_sample_notebook/A_generic_sample_notebook_to_perform_inference_on_ML_model_packages_from_AWS_Marketplace.ipynb

@@ -4,7 +4,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Deploy and perform inference on ML Model packages from AWS Marketplace.\n",
+    "# Deploy and perform inference on ML Model packages from AWS Marketplace\n",
     "\n",
     "There are two simple ways to try/deploy [ML model packages from AWS Marketplace](https://aws.amazon.com/marketplace/search/results?page=1&filters=FulfillmentOptionType%2CSageMaker::ResourceType&FulfillmentOptionType=SageMaker&SageMaker::ResourceType=ModelPackage), either using AWS console to deploy an ML model package (see [this blog](https://aws.amazon.com/blogs/machine-learning/adding-ai-to-your-applications-with-ready-to-use-models-from-aws-marketplace/)) or via code written typically in a Jupyter notebook. Many listings have a high-quality sample Jupyter notebooks provided by the seller itself, usually, these sample notebooks are linked to the AWS Marketplace listing (E.g. [Source Separation](https://aws.amazon.com/marketplace/pp/prodview-23n4vi2zw67we?qid=1579739476471&sr=0-1&ref_=srh_res_product_title)), If a sample notebook exists, try it out. \n",
     "\n",
@@ -14,7 +14,7 @@
     "\n",
     "> **Note**:If  you are facing technical issues while trying an ML model package from AWS Marketplace and need help, please open a support ticket or write to the team on [email protected] for additional assistance.\n",
     "\n",
-    "#### Pre-requisites:\n",
+    "## Pre-requisites\n",
     "1. Open this notebook from an Amazon SageMaker Notebook instance.\n",
     "1. Ensure that Amazon SageMaker notebook instance used  has  IAMExecutionRole with **AmazonSageMakerFullAccess**\n",
     "1. Your IAM role has these three permisions - **aws-marketplace:ViewSubscriptions**, **aws-marketplace:Unsubscribe**, **aws-marketplace:Subscribe** and you have authority to make AWS Marketplace subscriptions in the AWS account used.\n",
@@ -23,7 +23,7 @@
     "\n",
     "\n",
     "\n",
-    "#### Additional Resources:\n",
+    "## Additional Resources\n",
     "**Background on Model Packages**:\n",
     "1. An ML model can be created from a Model Package, to know how, see [Use a Model Package to Create a Model](https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-mkt-model-pkg-model.html). \n",
     "2. An ML Model accepts data and generates predictions.\n",
@@ -39,7 +39,7 @@
     "* For a Jupyter notebook of the sample solution for **Automating auto insurance claim processing workflow** outlined in [this re:Mars session](https://www.youtube.com/watch?v=GkKZt0s_ku0), see [amazon-sagemaker-examples/aws-marketplace](https://github.com/awslabs/amazon-sagemaker-examples/tree/master/aws_marketplace/using_model_packages/auto_insurance) GitHub repository.\n",
     "* For a Jupyter notebook of the sample solution for **Improving workplace safety solution** outlined in [this re:Invent session](https://www.youtube.com/watch?v=iLOXaWpK6ag), see [amazon-sagemaker-examples/aws-marketplace](https://github.com/awslabs/amazon-sagemaker-examples/tree/master/aws_marketplace/using_model_packages/improving_industrial_workplace_safety) GitHub repository.\n",
     "\n",
-    "#### Contents:\n",
+    "## Contents\n",
     "1. [Subscribe to the model package](#Subscribe-to-the-model-package)\n",
     "   1. [Identify compatible instance-type](#A.-Identify-compatible-instance-type)\n",
     "   2. [Identify content-type](#B.-Identify-content_type)\n",
@@ -57,7 +57,7 @@
     "4. [Delete the model](#4.-Delete-the-model)\n",
     "5. [Unsubscribe to the model package](#Unsubscribe-to-the-model-package)\n",
     "\n",
-    "#### Usage instructions\n",
+    "## Usage instructions\n",
     "You can run this notebook one cell at a time (By using Shift+Enter for running a cell)."
    ]
   },
@@ -112,7 +112,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 1. Subscribe to the model package"
+    "## 1. Subscribe to the model package"
    ]
   },
   {
@@ -140,7 +140,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### A. Identify compatible instance-type\n",
+    "### A. Identify compatible instance-type\n",
     "\n",
     "1. On the listing, Under **Pricing Information**, you will see **software pricing** for **real-time inference** as well as **batch-transform usage** for specific instance-types. \n",
     "\n",
@@ -168,7 +168,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### B. Identify content_type\n",
+    "### B. Identify content_type\n",
     "You need to specify input content-type and payload while performing inference on the model. In this sub-section you will identify input content type that is accepted by the model you wish to try. "
    ]
   },
@@ -214,7 +214,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### C. Specify model-package-arn\n",
+    "### C. Specify model-package-arn\n",
     "A model-package-arn is a unique identifier for each ML model package from AWS Marketplace within a chosen region."
    ]
   },
@@ -249,7 +249,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 2. Create an Endpoint and perform real-time inference."
+    "## 2. Create an Endpoint and perform real-time inference."
    ]
   },
   {
@@ -275,7 +275,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### A. Create an Endpoint"
+    "### A. Create an Endpoint"
    ]
   },
   {
@@ -304,7 +304,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### B. Create input payload"
+    "### B. Create input payload"
    ]
   },
   {
@@ -667,7 +667,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### C. Perform Real-time inference"
+    "### C. Perform Real-time inference"
    ]
   },
   {
@@ -739,7 +739,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### D. Visualize output"
+    "### D. Visualize output"
    ]
   },
   {
@@ -765,7 +765,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### E. Delete the endpoint"
+    "### E. Delete the endpoint"
    ]
   },
   {
@@ -789,7 +789,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 3. Perform Batch inference"
+    "## 3. Perform Batch inference"
    ]
   },
   {
@@ -838,7 +838,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### C. Visualize output"
+    "Visualize output"
    ]
   },
   {
@@ -879,7 +879,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 4. Delete the model"
+    "## 4. Delete the model"
    ]
   },
   {
@@ -902,7 +902,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### 5. Cleanup "
+    "## 5. Cleanup "
    ]
   },
   {