MATIH Platform is in active MVP development. Documentation reflects current implementation status.
15. Workbench Architecture
ML Workbench
Experiment Tracking

Experiment Tracking

The experiment tracking components provide MLflow-integrated experiment management with run comparison, metric visualization, and artifact inspection.

Components

ExperimentTracker (MLflow)

File: frontend/ml-workbench/src/components/MLflow/ExperimentTracker.tsx

MLflow experiment list and run management with metric filtering and comparison.

ExperimentTracker (Enhanced)

File: frontend/ml-workbench/src/components/ExperimentTracker/index.tsx

Platform-enhanced experiment tracker with advanced filtering, tagging, and team collaboration features.

ExperimentCompare

File: frontend/ml-workbench/src/components/advanced/ExperimentCompare.tsx

Side-by-side comparison of experiment runs with metric charts, parameter diffs, and artifact comparison.

MetricsVisualization

File: frontend/ml-workbench/src/components/advanced/MetricsVisualization.tsx

Interactive metric visualization with training curves, validation metrics, and custom metric plots using Recharts.

Model RegistryModel Deployment