MLFlow
tl; dr; A combinator governance component that provides a hosted MLFlow server, a model repository and experiment tracker.
Introduction
MLFlow is an open-source model registry and experiment tracker. Data scientists leverage this hosted service to store the results of their experiments and persist final artifacts like model parameters and weights.
This provides a centralised catalogue of models and experiments, which is useful for organizational purposes and sharing work.
MLFlow also comes with limited serving capabilities, although that is not it's core aim.
Test Drive
The fastest way to get started is to use the test drive functionality provided by TestFaster. Click on the "Launch Test Drive" button below (opens a new window).
Launch Test Drive Usage
Prerequisites
Start by preparing your Kubernetes cluster using one of the infrastructure components or use your own cluster.
Component Usage
module "mlflow" {
source = "combinator-ml/mlflow/k8s"
# Optional settings go here
}
See the full configuration options below.
Requirements
| Name | Version |
|---|---|
| terraform | >= 0.14 |
| helm | >= 2.0.0 |
| kubernetes | >= 2.0.0 |
Providers
| Name | Version |
|---|---|
| helm | >= 2.0.0 |
| kubernetes | >= 2.0.0 |
Modules
No Modules.
Resources
| Name |
|---|
| helm_release |
| kubernetes_secret |
Inputs
| Name | Description | Type | Default | Required |
|---|---|---|---|---|
| name_prefix | Prefix to be used when naming the different components. | string |
"combinator" |
no |
| namespace | The namespace to install into. | string |
"mlflow" |
no |
Outputs
No output.