Merge pull request #187327 from lgayhardt/patch-63

Update image syntax

Author: PRMerger13

articles/machine-learning/migrate-register-dataset.md

@@ -53,7 +53,7 @@ To download datasets directly:
 1. Select the dataset(s) you want to download.
 1. In the bottom action bar, select **Download**.
 
-    ![Screenshot showing how to download a dataset in Studio (classic)](./media/migrate-register-dataset/download-dataset.png)
+    :::image type="content" source="./media/migrate-register-dataset/download-dataset.png" alt-text="AScreenshot showing how to download a dataset in Studio (classic)." lightbox = "./media/migrate-register-dataset/download-dataset.png":::
 
 For the following data types, you must use the **Convert to CSV** module to download datasets.
 
@@ -73,7 +73,7 @@ To convert your dataset to a CSV and download the results:
 1. Right-click the **Convert to CSV** module.
 1. Select **Results dataset** > **Download**.
 
-    ![Screenshot showing how to setup a convert to CSV pipeline](./media/migrate-register-dataset/csv-download-dataset.png)
+    :::image type="content" source="./media/migrate-register-dataset/csv-download-dataset.png" alt-text="Screenshot showing how to setup a convert to CSV pipeline." lightbox = "./media/migrate-register-dataset/csv-download-dataset.png":::
 
 ### Upload your dataset to Azure Machine Learning
 
@@ -82,7 +82,8 @@ After you download the data file, you can register the dataset in Azure Machine
 1. Go to Azure Machine Learning studio ([ml.azure.com](https://ml.azure.com)).
 1. In the left navigation pane, select the **Datasets** tab.
 1. Select **Create dataset** > **From local files**.
-    ![Screenshot showing the datasets tab and the button for creating a local file](./media/migrate-register-dataset/register-dataset.png)
+
+    :::image type="content" source="./media/migrate-register-dataset/register-dataset.png" alt-text="Screenshot showing the datasets tab and the button for creating a local file.":::
 1. Enter a name and description.
 1. For **Dataset type**, select **Tabular**.