Merge pull request #89201 from j-martens/patch-652

Update tutorial-first-experiment-automated-ml.md

Author: ktoliver

articles/machine-learning/service/tutorial-first-experiment-automated-ml.md

@@ -64,22 +64,21 @@ You'll see the **Getting started** screen, since this is your first experiment w
 
 1. Enter this experiment name: `my-1st-automl-experiment`
 
-1. Select **Create a new compute**. A compute is a local or cloud based resource environment used to run your training script or host your service deployment. For this experiment we use a cloud based compute. 
+1. Select **Create a new compute** and configure your compute target. A compute target is a local or cloud based resource environment used to run your training script or host your service deployment. For this experiment we use a cloud based compute. 
 
-    1. Configure your compute context for this experiment.
-        
-        Field | Description | Value for tutorial
-        ----|---|---
-        Compute name |A unique name that identifies your compute context.|automl-compute
-        Virtual machine size| Select the virtual machine size for your compute.|Standard_DS12_V2
-        Min / Max nodes:| To enable data profiling, you must have one or more nodes.|Min nodes: 1<br>Max nodes: 6.
- 
-    1. To create your new compute, select **Create**. This takes a couple minutes to complete. 
+   Field | Description | Value for tutorial
+   ----|---|---
+   Compute name |A unique name that identifies your compute context.|automl-compute
+   Virtual&nbsp;machine&nbsp;size| Select the virtual machine size for your compute.|Standard_DS12_V2
+   Min / Max nodes (in Advanced Settings)| To profile data, you must specify 1 or more nodes.|Min nodes: 1<br>Max nodes: 6
 
-    1. When creation is complete, select your new compute from the drop-down list, and then select **Next**.
+   >[!NOTE]
+   >For this tutorial, you'll use the default storage account and container created with your new compute. They automatically populate in the form.
+    
+1. Select **Create** to get the compute target. 
+   **This takes a couple minutes to complete.** 
 
-    >[!NOTE]
-    >For this tutorial,  you'll use the default storage account and container created with your new compute. They automatically populate in the form.
+1. After creation, select your new compute target from the drop-down list and select **Next**.
 
 1. Select **Upload from local file** to begin creating a new dataset. 
 
@@ -107,23 +106,22 @@ You'll see the **Getting started** screen, since this is your first experiment w
 
         ![Preview tab configuration](media/tutorial-1st-experiment-automated-ml/schema-tab-config.gif)
 
-        
 1. Select **Classification** as the prediction task.
 
 1. Select **y** as the target column, what you want to predict. This column indicates whether the client subscribed to a term deposit or not.
 
 1. Expand **Advanced Settings** and populate the fields as follows.
 
-    Advanced settings|Value
-    ------|------
-    Primary metric| AUC_weighted 
-    Exit criteria| When any of these criteria are met, the training job ends before full completion: <br> *Training job time (minutes)*: 5  <br> *Max number of iterations*: 10 
-    Preprocessing| Enables preprocessing done by automated machine learning. This includes automatic data cleansing, preparing, and transformation to generate synthetic features.
-    Validation| Select K-fold cross-validation and **2** for the number of cross-validations. 
-    Concurrency| Select **5** for the number of max concurrent iterations.
-
    >[!NOTE]
    > For this experiment, you don't set a metric score or max cores per iterations threshold. You also don't block algorithms from being tested.
+   
+    Advanced&nbsp;settings|Description|Value&nbsp;for&nbsp;tutorial
+    ------|---------|---
+    Primary metric| |**AUC_weighted** 
+    Exit criteria| When any of these criteria are met, the training job ends even if it didn't fully complete. |Training&nbsp;job&nbsp;time&nbsp;(minutes): **5**  <br> <br> Max&nbsp;#&nbsp;of&nbsp;iterations&#58;**10** 
+    Preprocessing| Enables preprocessing done by automated machine learning. This includes automatic data cleansing, preparing, and transformation to generate synthetic features.| Enable
+    Validation| Validation type and number of tests. | **K-fold** cross-validation<br><br>  cross-validations: **2** 
+    Concurrency| The number of max concurrent iterations.|**5**
 
 1. Select **Start** to run the experiment. A screen appears with a status message as the experiment preparation begins.