Azure Machine Learning—Generally availability updates for November 2022
公開日: 11月 09, 2022
New features now available in GA include the ability to automate auto-shutdown/auto-start schedules and the functionality to configure and customize a compute instance in less time. You can also place customized tags on your workspaces to further monitor your Cost Analysis, seamlessly build NLP models with your own text data, easily build computer vision models without having to write any training code, and responsibly assess AI systems through versatile, end-to-end model assessments.
- Scheduled auto-shutdown/auto-start for compute instances: You can now optimize operating costs in an automated fashion using auto-shutdown and auto-start schedules.
- Automate compute instance customization and configuration: You can now streamline the process of setting up custom configurations, which reduces the time it takes to customize a compute instance.
- Assess AI systems and make data-driven decisions with the Responsible AI dashboard: You can now fully customize a variety of different Responsible AI components to generate a Responsible AI dashboard from a no-code UI experience.
- Build computer vision models using AutoML for Images: You can now completely control the building of your computer vision models for image classification, object detection, and instance segmentation.
- Build NLP models with AutoNLP capability: You can now build Natural Language Processing models (in 100+ languages) without needing to write and training code.