Integration guide
Connecting to the platform through training containers, DataOps, MLflow, the container registry and the REST API
This chapter is for people who connect to Geo-MLOps from outside the screens.
- External partners — you want to hand over your training code as a container image, or attach your own data server as a dataset source for the platform
- ML engineers — you want to log training you ran on a laptop to the platform, push images directly, or script repetitive work
How to use the screens is in the User guide. This chapter covers contracts (what is exchanged) and commands that actually work. Every command and piece of code in this chapter was run and checked on the capture stack (http://localhost:10000, tenant DEMO). In your environment, replace the address with something like https://mlops.example.com.
What do you want to do?
| I want to… | Read | Main readers |
|---|---|---|
| run my training code from the platform's training wizard | Build a training container | Partners, ML engineers |
| have the platform pull datasets from our data server | DataOps integration | Partners (data providers) |
| record training run on a laptop or in-house server on the platform's experiment screens | Log directly with MLflow | ML engineers |
use the platform registry with docker push/pull | Use the container registry directly | ML engineers, operators |
| automate uploads, queries and notifications with scripts | Automate with the REST API | ML engineers |
Three kinds of credentials
The platform uses a different token for each entry point. They are easy to mix up, so here they are first.
| Credential | Used for | Issued at | Tenant selection |
|---|---|---|---|
| Login session (cookie or JWT) | REST API (/api/v1/…) | POST /auth/cookie/login, POST /auth/jwt/login | X-Tenant header on every request |
| MLflow token | MLflow client (/mlflow) | Account settings (계정 설정) → MLflow tokens (MLflow 토큰) | Fixed in the token — no header needed |
| Container token | docker login (/v2) | Account settings (계정 설정) → Container tokens (컨테이너 토큰) | Fixed in the token — no header needed |
Inside a training container you do not need to create any of these yourself. The platform issues an MLflow token for each run, passes it in as an environment variable, and revokes it when the run ends.