ModelDB is an end-to-end system for managing machine learning models. It ingests models and associated metadata as models are being trained, stores model data in a structured format, and surfaces it through a web-frontend for rich querying.
- Kubernetes version 1.8+
- kubectl
- Helm
In this directory using Helm v3, run:
helm install --namespace <k8s namespace> <release-name> .
Alternatively, using Helm v2:
helm install . --name <release-name> --namespace <k8s namespace>
By default, the "default" namespace on your Kubernetes cluster is used.
Now that you have modelDB up and running on your K8s cluster, please visit our user guide and documentation to get started.
To build and deploy each of the services running as a part of modelDB, please follow the instructions in the corresponding service's repository to build the docker image for that service. Once the image is pushed to a container registry, update the corresponding property to point to the newly developed image in the values.yaml file.
To contribute to our project, look at the contributing section for each of the components -