此頁面尚未翻譯為您的語言。 請求翻譯

MLOE-11: Create tracking and version control mechanisms

PDFRSS

Due to its exploratory and iterative nature, it’s easy to lose track of ML model development and its evolution. You need to experiment with multiple combinations of data, algorithms, and parameters, all while observing the impact of incremental changes on model accuracy. Log and track your model experiments with configuration settings and hyperparameters. Document and version control any data processing-related findings, processes, and improvement to enable easier future referencing and reuse. Use a model registry to register and version control your ML models. Automate your model deployment with CI/CD processes. To learn more about knowledge management, refer the best practice documented in OPS11-BP04.

Implementation plan

• Track your ML experiments with SageMaker AI Experiments - Amazon SageMaker AI Experiments lets you create, manage, analyze, and compare your machine learning experiments. SageMaker AI Experiments automatically tracks the inputs, parameters, configurations, and results of your iterations as runs. You can assign, group, and organize these runs into experiments. SageMaker AI Experiments is integrated with Amazon SageMaker AI Studio, providing a visual interface to browse your active and past experiments, compare runs on key performance metrics, and identify the best performing models

• Associate notebook instances with Git repositories - To analyze data, document its processing, and evaluate ML models on Amazon SageMaker AI, you can use Amazon SageMaker AI Processing. SageMaker AI Processing can be used for feature engineering, data validation, model evaluation, and model interpretation. SageMaker AI notebook instances can be associated with Git repositories. This enables saving notebooks in a source control environment that persists after stopping or deleting the notebook instance. The notebooks hold the data processing code and its documentation. You can associate a default repository and up to three additional repositories with a notebook instance. The repositories can be hosted in GitHub or on any other Git server.

• Use SageMaker AI Model Registry - Catalog, manage, and deploy models using SageMaker AI Model Registry. Create a model group and, for each run of your ML pipeline, create a model version that you register in the model group.

Documents

• Amazon SageMaker AI – Process Data

• Associate Git Repositories with SageMaker AI Notebook Instances

Blogs

• Amazon SageMaker AI notebooks now support Git integration for increased persistence, collaboration, and reproducibility

• Track your ML experiments end to end with Data Version Control and Amazon SageMaker AI Experiments

Examples

• Amazon SageMaker AI processing