© Hipages Group Pty Ltd 2019

Helm Chart to deploy hyperopt: a Distributed Asynchronous Hyperparameter Optimization in Python. This repository uses the docker images published from hyperkops repository and provides a helm chart for deployment on Kubernetes.

$ helm install .

This chart bootstraps a Hyperopt deployment on a Kubernetes cluster using the Helm package manager.

• Kubernetes 1.4+ with Beta APIs enabled

First clone the repository using git clone:

git clone https://github.com/hipagesgroup/chart-hyperopt.git

To install the chart with the release name my-release:

$ helm install --name my-release .

The command deploys Hyperopt on the Kubernetes cluster in the default configuration. The configuration section lists the parameters that can be configured during installation.

Tip: List all releases using helm list

The helm chart outputs some useful information at the time of installation, like the portforward command required to connect to mongo db. This can be seen in the NOTES section of the output, eg.:

$ helm install .   

NAME:   silly-newt
LAST DEPLOYED: Tue Jun 25 10:50:15 2019
NAMESPACE: default
STATUS: DEPLOYED
...
...
...

NOTES:
1. To connect mongo db using your local to deploy hyperopt job

export POD_NAME=$(kubectl get pods --namespace default -l "app.kubernetes.io/name=chart-hyperopt-mongo,app.kubernetes.io/instance=silly-newt" -o jsonpath="{.items[0].metadata.name}")

kubectl port-forward $POD_NAME 27017:27017

You can then connect to mongo db at:
localhost:27017

2. you can connect to the mongo db on other deployments within the kubernetes environment using this service url:
"silly-newt-chart-hyperopt-mongo.default.svc.cluster.local"

To uninstall/delete the my-release deployment:

$ helm delete my-release

The command removes all the Kubernetes components associated with the chart and deletes the release.

A YAML file that specifies the values for the parameters can be provided while installing the chart. For example,

$ helm install --name my-release -f values.yaml .

Tip: You can use the default values.yaml but it is suggested you customise these values to suit your work load.

The following table lists the configurable parameters of the MongoDB chart and their default values.