Training Monitoring

48 public types in this namespace, organized by kind.

Models & Types (35)

Type Summary
ConfusionMatrixDataPoint Represents a confusion matrix data point.
ConsoleDashboard Console-based training dashboard with ASCII charts.
CurveDataPoint Represents a PR/ROC curve data point.
DatasetLineage Dataset lineage information.
DeploymentInfo Information about a model deployment.
EmailNotificationService Email notification service using SMTP.
EnvironmentInfo Environment information for reproducibility.
ExperimentInfo Information about an experiment.
ExperimentTracker Local file-based experiment tracker providing MLflow-compatible functionality.
GpuInfo GPU information.
HardwareInfo Hardware information.
HealthMonitorCallback<T> A ready-to-use ITrainingCallback that watches training health and aborts automatically when it detects a run that has gone wrong.
HistogramDataPoint Represents a histogram data point.
HtmlDashboard Generates interactive HTML dashboards for training visualization.
ImageDataPoint Represents an image data point.
LiveDashboard Provides a real-time training dashboard via embedded web server.
MetricValue A metric value with step information.
ModelLineage Model lineage information tracking how the model was created.
ModelRegistry Local file-based model registry for managing model versions and deployments.
ModelVersion A specific version of a registered model.
NotificationEventArgs Event arguments for notification events.
NotificationManager Manages multiple notification services and provides unified notification delivery.
RegisteredModel A registered model in the registry.
ResourceMonitor Monitors system resources (CPU, memory, GPU) during training.
ResourceSnapshot A snapshot of resource usage at a point in time.
ResourceThresholds Resource warning thresholds.
ResourceWarningEventArgs Event arguments for resource warnings.
RunComparison Comparison of multiple runs.
RunInfo Information about a run.
ScalarDataPoint Represents a scalar data point.
SlackNotificationService Slack notification service using webhooks or bot tokens.
SourceInfo Source information for a run.
TextDataPoint Represents a text data point.
TrainingMonitor<T> Implementation of training monitoring system for tracking model training progress.
TrainingNotification Represents a training notification.

Base Classes (1)

Type Summary
TrainingMonitorBase<T> Base class for training monitoring implementations.

Interfaces (4)

Type Summary
IExperimentTracker Interface for experiment tracking systems.
IModelRegistry Interface for a model registry that manages model versions and deployments.
INotificationService Interface for notification services.
ITrainingDashboard Interface for training dashboards that visualize metrics and training progress.

Enums (6)

Type Summary
DeploymentStatus Deployment status.
ModelStage Model version stage in the deployment lifecycle.
ModelVersionStatus Status of a model version.
NotificationSeverity Severity level for training notifications.
NotificationType Type of training notification.
RunStatus Status of an experiment run.

Options & Configuration (2)

Type Summary
EmailConfiguration Configuration for email notification service.
SlackConfiguration Configuration for Slack notification service.