We are excited to announce that Automated ML (AutoML) for Images within Azure Machine Learning (AzureML) is now generally available. AutoML for Images enables users to easily train computer vision models trained on image data, for tasks such as Image Classification. Object Detection and Instance Segmentation.
Image from: http://cs231n.stanford.edu/slides/2021/lecture_15.pdf
Customers across various industries are looking to leverage machine learning to build models that can process image data. Applications range from image classification of fashion photos to PPE detection in industrial environments. The ideal solution will allow users to easily build models, control the model training to optimize model performance, and offer a way to easily manage these ML models end-to-end. Data scientists have traditionally had to rely on the tedious process of custom training their image models. Iteratively finding the right set of model algorithms and hyperparameters for these scenarios typically requires significant time and effort.
With AutoML for Images, you can now easily build computer vision models without having to write any training code, while still maintaining complete control over model training, deployment and the e2e ML lifecycle of the model. This capability targets users with ML knowledge and it will boost data scientist productivity by offering the following capabilities -
- Ability to optimize model performance by controlling model algorithms + hyperparameters
- Control over model training / deployment environment
- Ability to deploy the model to the cloud or download for local use
- Seamless integration with AzureML Data Labeling
- Operationalization at scale with Azure Machine Learning’s MLOps
You can start creating AutoML models for computer vision tasks very easily, using either the Python SDK, CLI or the UI based AutoML job authoring experience in AzureML Studio.
The feature also allows you to use the following optional capabilities when building computer vision models -
- Support for Small Object Detection
- Incremental Training
- Automatic generation of ONNX and MLflow models
- Big data support: data streaming and multi-GPU/multi-node training support for all models
Learn how our customers are leveraging this capability in real world scenarios, to accelerate the end to end model building process in this customer story.
Summary
In summary, you can use AutoML for Images to easily build and optimize computer vision models, while maintaining flexibility and control over the entire model training and deployment process. Please give it a try and share your feedback with us.
Get Started Today!
- Watch our Azure Machine Learning breakout session
- Get started with Microsoft Learn to build skills
- Explore Azure Machine Learning announcements at Microsoft Ignite
- Read this Docs page to learn more about AutoML Image
Code examples
Review detailed code examples -
- Using the CLI in the azureml-examples repository for automated machine learning samples
- Using the Python SDK in the GitHub notebook repository for automated machine learning samples.