Creating your project - Amazon Lookout for Vision

Creating your project

An Amazon Lookout for Vision project is a grouping of the resources needed to create and manage a Lookout for Vision model. A project manages the following:

  • Dataset – The images and image labels used to train a model. For more information, see Creating your dataset.

  • Model – The software that you train to detect anomalies. You can have multiple versions of a model. For more information, see Training your model.

We recommend that you use a project for a single use case, such as detecting anomalies in a single type of machine part.

Note

You can use AWS CloudFormation to provision and configure Amazon Lookout for Vision projects. For more information, see Creating Amazon Lookout for Vision resources with AWS CloudFormation.

To view your projects, see Viewing your projects or open the Using the Amazon Lookout for Vision dashboard. To delete a model, see Deleting a model.

Creating a project (console)

The following procedure shows you how to create a project using the console.

To create a project (console)
  1. Open the Amazon Lookout for Vision console at https://console.aws.amazon.com/lookoutvision/.

  2. In the left navigation pane, choose Projects.

  3. Choose Create project.

  4. In Project name, enter a name for your project.

  5. Choose Create project. The details page for your project is displayed.

  6. Follow the steps in Creating your dataset to create your dataset.

Creating a project (SDK)

You use the CreateProject operation to create an Amazon Lookout for Vision project. The response from CreateProject includes the project name and the Amazon Resource Name (ARN) of the project. Afterwards, call CreateDataset to add a training and a test dataset to your project. For more information, see Creating a dataset with a manifest file (SDK).

To view the projects that you have created in a project, call ListProjects. For more information, see Viewing your projects.

To create a project (SDK)
  1. If you haven't already done so, install and configure the AWS CLI and the AWS SDKs. For more information, see Step 4: Set up the AWS CLI and AWS SDKs.

  2. Use the following example code to create a model.

    CLI

    Change the value of project-name to the name that you want to use for the project.

    aws lookoutvision create-project --project-name project name \ --profile lookoutvision-access
    Python

    This code is taken from the AWS Documentation SDK examples GitHub repository. See the full example here.

    @staticmethod def create_project(lookoutvision_client, project_name): """ Creates a new Lookout for Vision project. :param lookoutvision_client: A Boto3 Lookout for Vision client. :param project_name: The name for the new project. :return project_arn: The ARN of the new project. """ try: logger.info("Creating project: %s", project_name) response = lookoutvision_client.create_project(ProjectName=project_name) project_arn = response["ProjectMetadata"]["ProjectArn"] logger.info("project ARN: %s", project_arn) except ClientError: logger.exception("Couldn't create project %s.", project_name) raise else: return project_arn
    Java V2

    This code is taken from the AWS Documentation SDK examples GitHub repository. See the full example here.

    /** * Creates an Amazon Lookout for Vision project. * * @param lfvClient An Amazon Lookout for Vision client. * @param projectName The name of the project that you want to create. * @return ProjectMetadata Metadata information about the created project. */ public static ProjectMetadata createProject(LookoutVisionClient lfvClient, String projectName) throws LookoutVisionException { logger.log(Level.INFO, "Creating project: {0}", projectName); CreateProjectRequest createProjectRequest = CreateProjectRequest.builder().projectName(projectName) .build(); CreateProjectResponse response = lfvClient.createProject(createProjectRequest); logger.log(Level.INFO, "Project created. ARN: {0}", response.projectMetadata().projectArn()); return response.projectMetadata(); }
  3. Follow the steps in Creating a dataset using an Amazon SageMaker Ground Truth manifest file to create your dataset.