What MLOps Central Platform does
Features by menu, and the model lifecycle stage each menu handles.
After signing in you see the workspace menu on the left. The menu roughly follows the model lifecycle.
Feature map
| Stage | Menu | What it does |
|---|---|---|
| Data | Datasets (데이터셋) | Upload images, annotations, time series, tables and point clouds, or sync them from an external DataOps system |
| Training | Training (학습) | Pick the task, model, dataset, hyperparameters and runtime in an 11-step wizard and train on the GPU cluster |
| Tracking | Experiments (실험) | Every training run keeps its parameters, metric curves and artifacts. Select several runs to compare them |
| Versions | Model Registry (모델 레지스트리) | Manage model versions and stages (Staging, Production, Archived) and build serving images |
| Images | Container Registry (컨테이너 레지스트리) | Store training and serving container images. Import images built elsewhere as files |
| Review | Approvals (승인) | Approve or reject production promotion requests |
| Delivery | API Services (API 서비스) · Deployments (배포) | Serve a model as a REST API (try it in the inference console), or record deployments to field devices and roll them back |
| Field | Edge Fleet (Edge Fleet) · Collected data (수집 데이터) | Register devices, see health, collected data and inference results, send commands (such as pulling a model) and policies. Find data uploaded by every device in Collected data (수집 데이터) |
| Watch | Monitoring (모니터링) · Alerts (경보) · Drift (드리프트) | Key performance indicators (KPIs), alert rules and channels, input distribution shift and retraining |
| Automation | Pipelines (파이프라인) | Run the container image configured for each kind (Train, Register, Deploy, CT) as a Kubernetes Job and view the history |
Administration
Besides the regular menu there are two consoles.
- Admin console (tenant administrators): members and roles, MLflow and container tokens, audit log, training runner and dataset cache, and the tenant's execution environment such as training runtimes, dataset stagers and serving builders
- System console (global administrators): tenant creation, users, the global audit log, shared images, system health, build cache, weights mirror, Prometheus and Alertmanager, Kubernetes cluster, nodes, GPU slots, namespaces, workloads and storage
These are covered in Administration.
External integration
| What | How |
|---|---|
| Your own training code | Upload a container image that follows the contract and pick it in the training wizard |
| An in-house DataOps or labelling system | Implement three APIs to sync datasets |
| Experiment tracking from notebooks and scripts | Log with the standard MLflow client (issue a token) |
| Field devices | The geo-mlops-sdk agent handles collection, sync, model delivery and inference logging |
| Automation from other systems | REST API (OpenAPI documentation included) |
Details are in the Integration guide and Edge SDK.