Merge pull request #102815 from PeterCLu/plu-designer-tutorial

Update tutorial

Author: v-albemi

articles/machine-learning/media/tutorial-designer-automobile-price-deploy/create-inference-pipeline.png

@@ -35,6 +35,8 @@ To deploy your pipeline, you must first convert the training pipeline into a rea
 
 1. Above the pipeline canvas, select **Create inference pipeline** > **Real-time inference pipeline**.
 
+    ![Screenshot showing where to find the create pipeline button](./media/tutorial-designer-automobile-price-deploy/create-inference-pipeline.png)
+
     Your pipeline should now look like this: 
 
    ![Screenshot showing the expected configuration of the pipeline after preparing it for deployment](./media/tutorial-designer-automobile-price-deploy/real-time-inference-pipeline.png)

articles/machine-learning/media/tutorial-designer-automobile-price-deploy/real-time-inference-pipeline.png

@@ -9,13 +9,13 @@ services: machine-learning
 ms.service: machine-learning
 ms.subservice: core
 ms.topic: tutorial
-ms.date: 11/04/2019
+ms.date: 01/30/2020
 ---
 
 # Tutorial: Predict automobile price with the designer (preview)
 [!INCLUDE [applies-to-skus](../../includes/aml-applies-to-enterprise-sku.md)]
 
-In this two-part tutorial, you learn how to use the Azure Machine Learning designer to develop and deploy a predictive analytics solution that predicts the price of any car.
+In this two-part tutorial, you learn how to use the Azure Machine Learning designer to train and deploy a machine learning model that predicts the price of any car. The designer is a drag-and-drop tool that lets you create machine learning models without a single line of code.
 
 In part one of the tutorial, you'll learn how to:
 
@@ -41,13 +41,15 @@ To create an Azure Machine Learning pipeline, you need an Azure Machine Learning
 
 ### Create a new workspace
 
+In order to use the designer, you first need an Azure Machine Learning workspace. The workspace is the top-level resource for Azure Machine Learning, it provides a centralized place to work with all the artifacts you create in Azure Machine Learning.
+
 If you have an Azure Machine Learning workspace with an Enterprise edition, [skip to the next section](#create-the-pipeline).
 
 [!INCLUDE [aml-create-portal](../../includes/aml-create-in-portal-enterprise.md)]
 
 ### Create the pipeline
 
-1. Sign in to [ml.azure.com](https://ml.azure.com), and select the workspace you want to work with.
+1. Sign in to <a href="https://ml.azure.com?tabs=jre" target="_blank">ml.azure.com</a>, and select the workspace you want to work with.
 
 1. Select **Designer**.
 
@@ -57,6 +59,30 @@ If you have an Azure Machine Learning workspace with an Enterprise edition, [ski
 
 1. At the top of the canvas, select the default pipeline name **Pipeline-Created-on**. Rename it to *Automobile price prediction*. The name doesn't need to be unique.
 
+## Set the default compute target
+
+A pipeline runs on a compute target, which is a compute resource that's attached to your workspace. After you create a compute target, you can reuse it for future runs.
+
+You can set a **Default compute target** for the entire pipeline, which will tell every module to use the same compute target by default. However, you can specify compute targets on a per-module basis.
+
+1. Next to the pipeline name, select the **Gear icon** ![Screenshot of the gear icon](./media/tutorial-designer-automobile-price-train-score/gear-icon.png) at the top of the canvas to open the **Settings** pane.
+
+1. In the **Settings** pane to the right of the canvas, select **Select compute target**.
+
+    If you already have an available compute target, you can select it to run this pipeline.
+
+    > [!NOTE]
+    > The designer can run experiments only on Azure Machine Learning Compute targets. Other compute targets won't be shown.
+
+1. Enter a name for the compute resource.
+
+1. Select **Save**.
+
+    > [!NOTE]
+    > It takes approximately five minutes to create a compute resource. After the resource is created, you can reuse it and skip this wait time for future runs.
+    >
+    > The compute resource autoscales to zero nodes when it's idle to save cost. When you use it again after a delay, you might experience approximately five minutes of wait time while it scales back up.
+
 ## Import data
 
 There are several sample datasets included in the designer for you to experiment with. For this tutorial, use **Automobile price data (Raw)**. 
@@ -73,7 +99,7 @@ You can visualize the data to understand the dataset that you'll use.
 
 1. Select the **Automobile price data (Raw)** module.
 
-1. In the properties pane to the right of the canvas, select **Outputs**.
+1. In the module details pane to the right of the canvas, select **Outputs**.
 
 1. Select the graph icon to visualize the data.
 
@@ -89,9 +115,9 @@ Datasets typically require some preprocessing before analysis. You might have no
 
 ### Remove a column
 
-When you train a model, you have to do something about the data that's missing. In this dataset, the **normalized-losses** column is missing many values, so you exclude that column from the model altogether.
+When you train a model, you have to do something about the data that's missing. In this dataset, the **normalized-losses** column is missing many values, so you will exclude that column from the model altogether.
 
-1. Enter **Select** in the search box at the top of the palette to find the **Select Columns in Dataset** module.
+1. In the module palette to the left of the canvas, expand the **Data Transformation** section and find the **Select Columns in Dataset** module.
 
 1. Drag the **Select Columns in Dataset** module onto the canvas. Drop the module below the dataset module.
 
@@ -105,7 +131,7 @@ When you train a model, you have to do something about the data that's missing.
 
 1. Select the **Select Columns in Dataset** module.
 
-1. In the properties pane to the right of the canvas, select **All columns**.
+1. In the module details pane to the right of the canvas, select **All columns**.
 
 1. Select the **+** to add a new rule.
 
@@ -119,7 +145,7 @@ When you train a model, you have to do something about the data that's missing.
 
 1. Select the **Select Columns in Dataset** module. 
 
-1. In the properties pane, select the **Comment** text box and enter *Exclude normalized losses*.
+1. In the module details pane to the right of the canvas, select the **Comment** text box and enter *Exclude normalized losses*.
 
     Comments will appear on the graph to help you organize your pipeline.
 
@@ -130,13 +156,15 @@ Your dataset still has missing values after you remove the **normalized-losses**
 > [!TIP]
 > Cleaning the missing values from input data is a prerequisite for using most of the modules in the designer.
 
-1. Enter **Clean** in the search box to find the **Clean Missing Data** module.
+1. In the module palette to the left of the canvas, expand the section **Data Transformation**, and find the **Clean Missing Data** module.
 
 1. Drag the **Clean Missing Data** module to the pipeline canvas. Connect it to the **Select Columns in Dataset** module. 
 
-1. In the properties pane, select **Remove entire row** under **Cleaning mode**.
+1. Select the **Clean Missing Data** module.
+
+1. In the module details pane to the right of the canvas, select **Remove entire row** under **Cleaning mode**.
 
-1. In the properties pane **Comment** box, enter *Remove missing value rows*. 
+1. In the module details pane to the right of the canvas, select the **Comment** box, and enter *Remove missing value rows*. 
 
     Your pipeline should now look something like this:
 
@@ -152,26 +180,28 @@ Because you want to predict price, which is a number, you can use a regression a
 
 Splitting data is a common task in machine learning. You will split your data into two separate datasets. One dataset will train the model and the other will test how well the model performed.
 
-1. Enter **split data** in the search box to find the **Split Data** module. Connect the left port of the **Clean Missing Data** module to the **Split Data** module.
+1. In the module palette, expand the section **Data Transformation** and find the **Split Data** module.
+
+1. Drag the **Split Data** module to the pipeline canvas.
+
+1. Connect the left port of the **Clean Missing Data** module to the **Split Data** module.
 
     > [!IMPORTANT]
     > Be sure that the left output ports of **Clean Missing Data** connects to **Split Data**. The left port contains the the cleaned data. The right port contains the discarted data.
 
 1. Select the **Split Data** module.
 
-1. In the properties pane, set the **Fraction of rows in the first output dataset** to 0.7.
+1. In the module details pane to the right of the canvas, set the **Fraction of rows in the first output dataset** to 0.7.
 
     This option splits 70 percent of the data to train the model and 30 percent for testing it. The 70 percent dataset will be accessible through the left output port. The remaining data will be available through the right output port.
 
-1. In the properties pane **Comment** box, enter *Split the dataset into training set (0.7) and test set (0.3)*.
+1. In the module details pane to the right of the canvas, select the **Comment** box, and enter *Split the dataset into training set (0.7) and test set (0.3)*.
 
 ### Train the model
 
 Train the model by giving it a dataset that includes the price. The algorithm constructs a model that explains the relationship between the features and the price as presented by the training data.
 
-1. To select the learning algorithm, clear your module palette search box.
-
-1. Expand **Machine Learning Algorithms**.
+1. In the module palette, expand **Machine Learning Algorithms**.
 
     This option displays several categories of modules that you can use to initialize learning algorithms.
 
@@ -188,9 +218,11 @@ Train the model by giving it a dataset that includes the price. The algorithm co
 
     ![Screenshot showing the correct configuration of the Train Model module. The Linear Regression module connects to left port of Train Model module and the Split Data module connects to right port of Train Model](./media/tutorial-designer-automobile-price-train-score/pipeline-train-model.png)
 
+1. In the module palette, expand the section **Module training**, and drag the **Train Model** module to the canvas.
+
 1. Select the **Train Model** module.
 
-1. In the properties pane, select **Edit column** selector.
+1. In the module details pane to the right of the canvas, select **Edit column** selector.
 
 1. In the **Label column** dialog box, expand the drop-down menu and select **Column names**. 
 
@@ -200,15 +232,15 @@ Train the model by giving it a dataset that includes the price. The algorithm co
 
     ![Screenshot showing the correct configuration of the pipeline after adding the Train Model module.](./media/tutorial-designer-automobile-price-train-score/pipeline-train-graph.png)
 
-## Score a machine learning model
+### Add the Score Model module
 
 After you train your model by using 70 percent of the data, you can use it to score the other 30 percent to see how well your model functions.
 
 1. Enter *score model* in the search box to find the **Score Model** module. Drag the module to the pipeline canvas. 
 
 1. Connect the output of the **Train Model** module to the left input port of **Score Model**. Connect the test data output (right port) of the **Split Data** module to the right input port of **Score Model**.
 
-## Evaluate a machine learning model
+### Add the Evaluate Model module
 
 Use the **Evaluate Model** module to evaluate how well your model scored the test dataset.
 
@@ -222,15 +254,28 @@ Use the **Evaluate Model** module to evaluate how well your model scored the tes
 
 ## Run the pipeline
 
-[!INCLUDE [aml-ui-create-training-compute](../../includes/aml-ui-create-training-compute.md)]
+Now that your pipeline is all setup, you can submit a pipeline run.
+
+1. At the top of the canvas, select **Run**.
+
+1. In the **Set up pipeline run** dialog box, select **+ New experiment** for the **Experiment**.
+
+    > [!NOTE]
+    > Experiments group similar pipeline runs together. If you run a pipeline multiple times, you can select the same experiment for successive runs.
+
+    1. Enter a descriptive name for **Experiment Name**.
+
+    1. Select **Run**.
+    
+    You can view run status and details at the top right of the canvas.
 
 ### View scored labels
 
 After the run completes, you can view the results of the pipeline run. First, look at the predictions generated by the regression model.
 
 1. Select the **Score Model** module to view its output.
 
-1. In the properties pane, select **Outputs** > graph icon ![visualize icon](./media/tutorial-designer-automobile-price-train-score/visualize-icon.png) to view results.
+1. In the module details pane to the right of the canvas, select **Outputs** > graph icon ![visualize icon](./media/tutorial-designer-automobile-price-train-score/visualize-icon.png) to view results.
 
     Here you can see the predicted prices and the actual prices from the testing data.
 
@@ -242,7 +287,7 @@ Use the **Evaluate Model** to see how well the trained model performed on the te
 
 1. Select the **Evaluate Model** module to view its output.
 
-1. In the properties pane, select **Output** > graph icon ![visualize icon](./media/tutorial-designer-automobile-price-train-score/visualize-icon.png) to view results.
+1. In the module details pane to the right of the canvas, select **Output** > graph icon ![visualize icon](./media/tutorial-designer-automobile-price-train-score/visualize-icon.png) to view results.
 
 The following statistics are shown for your model:
 
@@ -256,17 +301,12 @@ For each of the error statistics, smaller is better. A smaller value indicates t
 
 ## Clean up resources
 
+Skip this section if you want to continue on with part 2 of the tutorial, [deploying models](tutorial-designer-automobile-price-deploy.md).
+
 [!INCLUDE [aml-ui-cleanup](../../includes/aml-ui-cleanup.md)]
 
 ## Next steps
 
-In part one of this tutorial, you completed the following tasks:
-
-* Create a pipeline
-* Prepare the data
-* Train the model
-* Score and evaluate the model
-
 In part two, you'll learn how to deploy your model as a real-time endpoint.
 
 > [!div class="nextstepaction"]