ContextQMD
Back to Aws Doc Sdk Examples

Lookout for Vision code examples for the SDK for Java 2.x

javav2/example_code/lookoutvision/README.md

latest11.3 KB
Original Source

Lookout for Vision code examples for the SDK for Java 2.x

Overview

Shows how to use the AWS SDK for Java 2.x to work with Amazon Lookout for Vision.

<!--custom.overview.start--> <!--custom.overview.end-->

Lookout for Vision enables you to find visual defects in industrial products, accurately and at scale.

⚠ Important

  • Running this code might result in charges to your AWS account. For more details, see AWS Pricing and Free Tier.
  • Running the tests might result in charges to your AWS account.
  • We recommend that you grant your code least privilege. At most, grant only the minimum permissions required to perform the task. For more information, see Grant least privilege.
  • This code is not tested in every AWS Region. For more information, see AWS Regional Services.
<!--custom.important.start--> <!--custom.important.end-->

Code examples

Prerequisites

For prerequisites, see the README in the javav2 folder.

<!--custom.prerequisites.start--> <!--custom.prerequisites.end--> <!--custom.examples.start-->

Example code structure

The following files provide utility classes for managing Amazon Lookout for Vision resources.

  • Models.java - A class of static functions that manage Amazon Lookout for Vision models.
  • Projects.java - A class of static functions that manage Amazon Lookout for Vision projects.
  • Datasets.java - A class of static functions that manage Amazon Lookout for Vision datasets.
  • Hosting.java - A class of static functions that manage the hosting of Amazon Lookout for Vision models.
  • EdgePackages.java - A class of static functions that manage Amazon Lookout for Vision edge packaging jobs.

The following examples use the utility classes to show how to use the Amazon Lookout for Vision API.

  • CreateDataset.java - Shows how to create an Amazon Lookout for Vision dataset with CreateDataset. You must have a manifest file to train the model. We provide a script that creates a manifest file from a .csv file. For more information, see Creating a manifest file.

  • CreateModel.java - Shows how to create an Amazon Lookout for Vision model with CreateModel. You are charged for the amount of time it takes to successfully train a model.

  • CreateProject.java - Shows how to create an Amazon Lookout for Vision project with CreateProject.

  • DeleteDataset.java - Shows how to delete an Amazon Lookout for Vision dataset with DeleteDataset.

  • DeleteModel.java - Shows how to delete an Amazon Lookout for Vision model with DeleteModel.

  • DeleteProject.java - Shows how to delete an Amazon Lookout for Vision project with DeleteProject.

  • DescribeDataset.java - Shows how to get information about an Amazon Lookout for Vision dataset with DescribeDataset.

  • DescribeModel.java - Shows how to get information about an Amazon Lookout for Vision model with DescribeModel.

  • DescribeModelPackagingJob.java - Shows how to get information about an Amazon Lookout for Vision model packaging job with DescribeModelPackagingJob.

  • DescribeProject.java - Shows how to get information about an Amazon Lookout for Vision project with DescribeProject.

  • DetectAnomalies.java - Shows how to analyze an image for anomalies with DetectAnomalies. The configuration file is is analysis-config.json in the resources folder. For more information, see Determining if an image is anomalous.

  • ListDatasetEntries.java - Shows how to list the JSON lines in an Amazon Lookout for Vision dataset with ListDatasetEntries.

  • ListModelPackagingJobs.java - Shows how to list the Amazon Lookout for Vision model packaging jobs in a project with ListModelPackagingJobs.

  • ListModelTags.java - Shows how to list tags attached to an Amazon Lookout for Vision model with ListTagsForResource.

  • ListModels.java - Shows how to list the Amazon Lookout for Vision models in a project with ListModels.

  • ListProjects.java - Shows how to list the Amazon Lookout for Vision projects in the current AWS account and AWS Region with ListProjects.

  • ShowAnomalies.java - Shows how to display classification and segmentation (image masks and anomaly labels) information on an image analyzed by DetectAnomalies. For more information, see Showing classification and segmentation.

  • StartModel.java - Shows how to start hosting an Amazon Lookout for Vision model with StartModel. You are charged for the amount of time that your model is hosted.

  • StartModelPackagingJob.java - Shows how to start a Amazon Lookout for Vision model packaging job with StartModelPackagingJob. You specify the model packaging job settings in a JSON format file. We provide a template JSON file for a target device (Jetson Xavier) and a template JSON file for a target platform. For information about the package settings that you can make, see Packaging your model (SDK).

  • StopModel.java - Shows how to stop a hosted Amazon Lookout for Vision model with StopModel.

  • TagModel.java - Shows how to attach a tag to an Amazon Lookout for Vision model with TagResource.

  • UntagModel.java - Shows how to remove a tag from an Amazon Lookout for Vision model with UntagResource.

  • UpdateDatasetEntries.java - Shows how to update Amazon Lookout for Vision dataset with a manifest file with UpdateDatasetEntries.

<!--custom.examples.end-->

Run the examples

Instructions

<!--custom.instructions.start--> <!--custom.instructions.end-->

Tests

⚠ Running tests might result in charges to your AWS account.

To find instructions for running these tests, see the README in the javav2 folder.

<!--custom.tests.start-->

Properties file

Before running the Amazon Lookout for Vision JUnit tests, you must define values in the config.properties file located in the resources folder. This file contains values that are required to run the JUnit tests. For example, you define an instance name used for various tests. If you do not define all values, the JUnit tests fail.

Define these values to successfully run the JUnit tests:

  • projectName - The name of the project to create in the tests.
  • modelDescription - A description for the model.
  • modelTrainingOutputBucket - The Amazon S3 bucket in which to place the training results.
  • modelTrainingOutputFolder - The folder in modelTrainingOutputBucket in which to place the training results.
  • photo - The location of an image to analyze with the trained model.
  • anomalousPhoto = The location of an anomalous image to analyze with the trained model.
  • anomalyLabel = The label for an anomaly in the project.
  • manifestFile - The location of a local manifest file that is used to populate the training dataset. For more information, see Creating a manifest file.
  • modelPackageJobJsonFile - The location of the edge packaging Job request JSON file. We provide a template JSON file for a target device (Jetson Xavier) and a template JSON file for a target platform. To successfully run the model packaging job test, make sure that the value of ModelVersion is "1". Each time you run the test, you must change the value of ComponentVersion and JobName. For information about the package settings that you can make, see Packaging your model (SDK).

If you want to use the project, dataset, and model that the testing creates, disable the following tests:

  • deleteDataset_thenNotFound()
  • deleteModel_thenNotFound()
  • deleteProject_thenNotFound()
<!--custom.tests.end-->

Additional resources

<!--custom.resources.start--> <!--custom.resources.end-->

Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.

SPDX-License-Identifier: Apache-2.0