fix

Author: v-thepet

articles/machine-learning/concept-machine-learning-registries-mlops.md

@@ -31,16 +31,16 @@ In the preceding scenarios, you might use different Azure Machine Learning works
 
 A registry, much like a Git repository, decouples machine learning assets from workspaces and hosts the assets in a central location, making them available to all workspaces in your organization.
 
-To promote models across development, test, and production environments, start by iteratively developing a model in the development environment. When you have a good candidate model, you can publish it to a registry. You can then deploy the model from the registry to endpoints in different workspaces.
+To promote models across development, test, and production environments, you can start by iteratively developing a model in the development environment. When you have a good candidate model, you can publish it to a registry. You can then deploy the model from the registry to endpoints in different workspaces.
 
 > [!TIP]
 > If you already have models registered in a workspace, you can promote the models to a registry. You can also register a model directly in a registry from the output of a training job.
 
-To develop a pipeline in one workspace and then run it in other workspaces, start by registering the components and environments that form the building blocks of the pipeline. When you submit the pipeline job, the workspace to run in is determined by the compute and the training data, which are unique to each workspace.
+To develop a pipeline in one workspace and then run it in other workspaces, start by registering the components and environments that form the building blocks of the pipeline. When you submit the pipeline job, the compute and the training data, which are unique to each workspace, determine the workspace to run in.
 
 The following diagram shows training pipeline promotion between exploratory and development workspaces, then trained model promotion to test and production.
 
-:::image type="content" source="./media/concept-machine-learning-registries-mlops/cross-workspace-mlops-with-registries.png" alt-text="Diagram of pipeline and model use across environments.":::
+:::image type="content" source="./media/concept-machine-learning-registries-mlops/cross-workspace-mlops-with-registries.png" alt-text="Diagram of pipeline and model use across environments." border="false":::
 
 ## Next steps