II.
LibraryProcess overview
Reference · livelib-process:data-science-ml--ml-integration-testing
ml-integration-testing overview
ML System Integration Testing - Validate end-to-end ML pipeline integration across data ingestion, preprocessing, model training, serving, and monitoring components with quality gates and validation loops.
Attributes
displayName
ml-integration-testing
description
ML System Integration Testing - Validate end-to-end ML pipeline integration across data ingestion,
preprocessing, model training, serving, and monitoring components with quality gates and validation loops.
libraryPath
library/specializations/data-science-ml/ml-integration-testing.js
specialization
data-science-ml
references
- - ML Testing: A Guide: https://madewithml.com/courses/mlops/testing/ - Google ML Testing Best Practices: https://developers.google.com/machine-learning/testing-debugging - AWS ML Testing: https://aws.amazon.com/blogs/machine-learning/testing-approaches-for-amazon-sagemaker-ml-models/ - Microsoft ML Testing: https://learn.microsoft.com/en-us/azure/architecture/guide/testing/mission-critical-deployment-testing - Integration Testing Patterns: https://martinfowler.com/articles/microservice-testing/ - ML Observability: https://neptune.ai/blog/ml-model-testing
example
const result = await orchestrate('specializations/data-science-ml/ml-integration-testing', {
systemName: 'Recommendation Engine',
components: ['data-pipeline', 'feature-store', 'model-service', 'monitoring'],
testEnvironment: 'staging',
integrationScenarios: [
{ name: 'end-to-end-prediction', type: 'e2e' },
{ name: 'model-update-rollout', type: 'deployment' },
{ name: 'data-drift-detection', type: 'monitoring' }
],
performanceRequirements: {
latency: { p95: 100, p99: 200 },
throughput: { min: 1000 },
accuracy: { min: 0.85 }
},
targetCoverage: 85
});
usesAgents
- general-purpose
Outgoing edges
lib_applies_to_domain1
- domain:data-science·DomainData Science
lib_belongs_to_specialization1
- specialization:data-science-ml·Specialization
lib_implements_workflow2
- workflow:code-review·Workflow
- workflow:ml-model-lifecycle·WorkflowML Model Lifecycle
Incoming edges
None.