ChurnWatch
An MLOps practice project built on the AWS ecosystem, used to demonstrate a complete data drift monitoring workflow covering model training, PSI drift detection and CloudWatch metric monitoring.
Timeline (last 90 days)
ChurnWatch is an MLOps project based on the AWS ecosystem, covering the complete pipeline of model training, drift detection, and metric monitoring.
ChurnWatch uses Amazon S3 to store training datasets, Amazon SageMaker to train the churn prediction model, Python scripts to perform drift detection with PSI, and Amazon CloudWatch for metric monitoring.
ChurnWatch covers both ends of the MLOps lifecycle, training and monitoring, omitting heavy components such as CI/CD, feature stores, and model registries.
Extending ChurnWatch can introduce Amazon SageMaker Model Monitor for automatic drift detection, use SageMaker Pipelines to orchestrate the retraining process, and use Feature Store to unify feature definitions, thereby connecting detected drift to a closed loop of automatically triggered retraining.
All Facts (4)
ChurnWatch is an MLOps project based on the AWS ecosystem, covering the complete pipeline of model training, drift detection, and metric monitoring.
50%UnverifiedChurnWatch uses Amazon S3 to store training datasets, Amazon SageMaker to train the churn prediction model, Python scripts to perform drift detection with PSI, and Amazon CloudWatch for metric monitoring.
50%UnverifiedChurnWatch covers both ends of the MLOps lifecycle, training and monitoring, omitting heavy components such as CI/CD, feature stores, and model registries.
50%UnverifiedExtending ChurnWatch can introduce Amazon SageMaker Model Monitor for automatic drift detection, use SageMaker Pipelines to orchestrate the retraining process, and use Feature Store to unify feature definitions, thereby connecting detected drift to a closed loop of automatically triggered retraining.
50%