This page guides you through cleaning up the Google Cloud resources that youcreated to train your image classification model and serve predictions from it.

1. Setting up your project and environment.

2. Training a custom image classificationmodel.

3. Serving predictions from a custom image classificationmodel.

4. Cleaning up your project.

Each page assumes that you have already performed the instructions from theprevious pages of the tutorial.

1. In the Google Cloud console, activate Cloud Shell.

   Activate Cloud Shell

2. In the Cloud Shell session, run the following command:

   cdhello-custom-sample

Delete Vertex AI resources

This section describes how to delete all of the Vertex AI resourcesthat you created for this tutorial.

Undeploy your model from your endpoint

This section describes how to undeploy your model from your endpoint. You canthink about this action as a way of disconnecting your model from your endpoint.

You must follow this section before you candelete yourendpoint ordelete your model.

1. In the Google Cloud console, in the Vertex AI section, go totheEndpoints page.

   Go to Endpoints

2. Clickhello_custom to go to the endpoint details page.

3. On the row for your model,hello_custom, clickUndeploy modeldelete.

4. In theUndeploy model from endpoint dialog, clickUndeploy.

Delete your endpoint

Before you delete an endpoint, you mustundeploy your model from yourendpoint. After you've deleted your endpoint, you won'tbe able to re-use that endpoint name for up to 7 days.

After you've undeployed your model from the endpoint, do the followingto delete your endpoint:

1. In the Google Cloud console, in the Vertex AI section, go totheEndpoints page.

   Go to Endpoints

2. Find your the row of your endpoint,hello_custom, again. On that row, clickView moremore_vert. Then clickRemove endpoint.

3. In theRemove endpoint dialog, clickConfirm.

Delete your model

Before you follow this section, you mustundeploy your model from yourendpoint. Afterward, do the following to delete your model:

1. In the Google Cloud console, in the Vertex AI section, go totheModels page.

   Go to Models

2. Find your the row of your model,hello_custom. On that row, clickViewmoremore_vert. ThenclickDelete model.

3. In theDelete model dialog, clickDelete.

Delete your custom training pipeline and job

Your training pipeline and custom job are just records of the training thathappened earlier. If you want to delete your custom job, do the following:

1. In the Google Cloud console, in the Vertex AI section, go totheTraining pipelines page.

   Go to Training pipelines

2. Find your the row of your training pipeline,hello_custom. On that row,clickView moremore_vert. Then clickDelete trainingpipeline.

3. In theDelete training job dialog, clickDelete.

4. To go to theCustom jobs page, clickCustom job in theGoogle Cloud console, or click the following link:

   Go to Custom jobs

5. Find your the row of your custom job,hello_custom-custom-job. On that row,clickView moremore_vert. Then clickDelete custom job.

6. In theDelete training job dialog, clickDelete.

Clean up your Cloud Shell session

Cloud Shell incurs no charges, and itautomatically deletes your homedisk after a period of inactivity. However, if youplan to use Cloud Shell for other purposes in the near future, youmight want to manually remove the files that you created for this tutorial.

In your Cloud Shell session, run the following commands:

cd..rm-rfhello-custom-sample

Delete your Cloud Storage bucket

In your Cloud Shell session, run the following command:

gcloudstoragermgs://BUCKET_NAME--recursive--continue-on-error

ReplaceBUCKET_NAME with the name of the Cloud Storagebucket that you created when reading thefirst page of thistutorial.

Delete your Cloud Run function

In your Cloud Shell session, run the following command:

gcloudfunctionsdeleteclassify_flower--region=us-central1--quiet

What's next

• To learn about additional ways to train ML models on Vertex AI,try one of the otherVertex AI tutorials.

• Read anoverview of how Vertex AIworks.