Fix bookmark reference bugs . . .

Author: fbsolo-ms1

articles/machine-learning/how-to-manage-inputs-outputs-pipeline.md

@@ -297,7 +297,7 @@ az ml job download --output-name <OUTPUT_PORT_NAME> -n <JOB_NAME> -g <RESOURCE_G
 ```
 # [Python SDK](#tab/python)
 
-Before we dive in the code, you need a way to reference your workspace. You create `ml_client` for a handle to the workspace. Refer to [Create handle to workspace](./tutorial-explore-data.md#create-handle-to-workspace) to initialize `ml_client`.
+Before we dive in the code, you need a way to reference your workspace. You create `ml_client` for a handle to the workspace. Refer to [Create handle to workspace](./tutorial-explore-data.md#create-a-handle-to-the-workspace) to initialize `ml_client`.
 
 ```python
 # Download all the outputs of the job
@@ -325,7 +325,7 @@ az ml job download --all -n <JOB_NAME> -g <RESOURCE_GROUP_NAME> -w <WORKSPACE_NA
 
 # [Python SDK](#tab/python)
 
-Before we dive in the code, you need a way to reference your workspace. You create `ml_client` for a handle to the workspace. Refer to [Create handle to workspace](./tutorial-explore-data.md#create-handle-to-workspace) to initialize `ml_client`.
+Before we dive in the code, you need a way to reference your workspace. You create `ml_client` for a handle to the workspace. Refer to [Create handle to workspace](./tutorial-explore-data.md#create-a-handle-to-the-workspace) to initialize `ml_client`.
 
 ```python
 # List all child jobs in the job