We are making the following features generally available in Azure Machine Learning:
- Data Labeling-Image Instance Segmentation in Azure Machine Learning gives you a central place to create, manage, and monitor labeling projects. Use it to coordinate data, labels, and team members to efficiently manage labeling tasks. It supports image classification, either multi-label or multi-class, object identification with bounded boxes, and Image Instance Segmentation (polygon).
- MLflow Support: Azure Machine Learning supports managing the end-to-end machine learning cycle using open MLFlow standards, enabling existing workloads to move from local execution to execution in the intelligent cloud & edge.
-
Use MLflow tracking and logging API’s
-
Track experiments and metrics with both services
-
Track models in the MLflow and Azure Machine Learning model registry
-
Deploy MLflow tracked models using the azureml-mlflow package to ACI and AKS
Learn more on how to label images.
Read the how-to doc on MLflow.