Merge pull request #49991 from kgremban/aug24-mlfix

Update ML doc for IoT Edge GA

Author: TrishaMc

articles/machine-learning/desktop-workbench/deploy-to-iot-edge-device.md

@@ -10,31 +10,31 @@ ms.service: machine-learning
 ms.component: core
 ms.workload: data-services
 ms.topic: article
-ms.date: 2/1/2018
+ms.date: 08/24/2018
 ---
 # Deploy an Azure Machine Learning model to an Azure IoT Edge device
 
-All Azure Machine Learning models containerized as Docker-based web services can also run on Azure IoT Edge devices. Additional scripts and instructions can be found in the [AI Toolkit for Azure IoT Edge](http://aka.ms/AI-toolkit).
+Azure Machine Learning models can be containerized as Docker-based web services. Azure IoT Edge enables you to deploy containers remotely onto devices. Use these services together to run your models at the edge for faster response times and less data transfer. 
+
+Additional scripts and instructions can be found in the [AI Toolkit for Azure IoT Edge](http://aka.ms/AI-toolkit).
 
 ## Operationalize the model
+
+Azure IoT Edge modules are based on container images. To deploy your Machine Learning model to an IoT Edge device, you need to create a Docker image.
+
 Operationalize your model by following the instructions in [Azure Machine Learning Model Management Web Service Deployment](model-management-service-deploy.md) to create a Docker image with your model.
 
 ## Deploy to Azure IoT Edge
-Azure IoT Edge moves cloud analytics and custom business logic to devices. All Machine Learning models can run on IoT Edge devices. The documentation to set up an IoT Edge device and create a deployment can be found at [aka.ms/azure-iot-edge-doc](https://aka.ms/azure-iot-edge-doc).
 
-The following are additional things to note.
+Once you have the image of your model, you can deploy it to any Azure IoT Edge device. All Machine Learning models can run on IoT Edge devices. 
+
+### Set up an IoT Edge device
 
-### Add registry credentials to the Edge runtime on your Edge device
-On the machine where you're running IoT Edge, add the credentials of your registry so the runtime can have access to pull the container.
+Use the Azure IoT Edge documentation to prepare a device. 
+
+1. [Register a device with Azure IoT Hub](../../iot-edge/how-to-register-device-portal.md). The output of this processes is a connection string that you can use to configure your physical device. 
+2. Install the IoT Edge runtime on your physical device, and configure it with a connection string. You can install the runtime on [Windows](../../iot-edge/how-to-install-iot-edge-windows-with-windows.md) or [Linux](../../iot-edge/how-to-install-iot-edge-linux.md) devices.  
 
-For Windows, run the following command:
-```cmd/sh
-iotedgectl login --address <docker-registry-address> --username <docker-username> --password <docker-password>
-```
-For Linux, run the following command:
-```cmd/sh
-sudo iotedgectl login --address <docker-registry-address> --username <docker-username> --password <docker-password>
-```
 
 ### Find the Machine Learning container image location
 You need the location of your Machine Learning container image. To find the container image location:
@@ -43,7 +43,17 @@ You need the location of your Machine Learning container image. To find the cont
 2. In the **Azure Container Registry**, select the registry you wish to inspect.
 3. In the registry, click **Repositories** to see a list of all the repositories and their images.
 
+While you're looking at your container registry in the Azure portal, retrieve the container registry credentials. These credentials need to be given to the IoT Edge device so that it can pull the image from your private registry. 
+
+1. In the container registry, click **Access keys**. 
+2. **Enable** the admin user, if it isn't already. 
+3. Save the values for **Login server**, **Username**, and **password**. 
+
+### Deploy the container image to your device
+
+With the container image and the container registry credentials, you're ready to deploy the machine learning model to your IoT Edge device. 
 
+Follow the instructions in [Deploy IoT Edge modules from the Azure portal](../../iot-edge/how-to-deploy-modules-portal.md) to launch your model on your IoT Edge device.