Public preview: Azure Machine Learning updates
公開日: 5月 24, 2022
Azure Machine Learning's Responsible AI (RAI) dashboard, now in preview, is a new feature that helps developers and data scientists implement Responsible AI easily. The dashboard brings together multiple capabilities such as data explorer, fairness, model interpretability, error analysis, counterfactual and causal inference analysis, which help developers debug their models and make more informed data-driven decisions. In addition, Azure Machine Learning now offers a Responsible AI scorecard to summarize model performance and insights, helping technical and non-technical audiences understand the impact of applying Responsible AI.
Additional preview updates:
- AutoML features include support for natural language processing and image tasks, generation of models’ training codes, and enhancements for product integration and Machine Learning Operations (MLOps).
- The Python SDK v2 simplifies the developer experience.
- RStudio is one of the most popular integrated development environments among R developers for machine learning and data science projects. Azure Machine Learning now features a custom application that sets up RStudio Workbench into your Azure Machine Learning environment when an RStudio Workbench license key is provided. RStudio Workbench also provides access to other development environments, including Jupyter Notebooks and VSCode.
Learn more about these updates.