Hi <!channel>! BentoML 1.0.0a7 has just been released with a number of improvements and bug fixes. The two most significant changes are: • BREAKING CHANGE: Default serving port has been changed to 3000 ◦ This is due to an issue with new MacOS where 5000 port is always in use. ◦ This will affect default serving port when deploying with Docker. Existing 1.0 preview release users will need to either change deployment config to use port 3000, or pass

--port 5000

to the container command, in order to use the previous default port setting. • New import/export API is available now! ◦ Users can now export models and bentos from local store to a standalone file ◦ Lean more via

bentoml export --help

and

bentoml models export --help

We’ve also recently released Yatai version 0.2.1, with major refactoring around the deployment controller. Users can now create Bento deployments directly via

kubectl

and Kubernetes resource YAML file, in addition to the Yatai Web UI and REST API:

# my_deployment.yaml
apiVersion: <http://serving.yatai.ai/v1alpha1|serving.yatai.ai/v1alpha1>
kind: BentoDeployment
metadata:
  name: demo
spec:
  bento_tag: iris_classifier:3oevmqfvnkvwvuqj
  resources:
    limits:
      cpu: 1000m
    requests:
      cpu: 500m

Apply deployment to your cluster:

kubeclt apply -f my_deployment.yaml

This will make it easy for DevOps to customize BentoML deployments on Kubernetes cluster, with additional k8s resources such as credentials, db, policies and other services. On the

bentoctl

project, we are working on a major new version which embraces a workflow based on terraform, to simplify deploying Bentos to any cloud platforms, such as AWS EC2, Lambda, Sagemaker, Azure, GCP, Heroku, etc. If you are interested in learning more or help with beta testing, definitely chat with @Bo and @jjmachan.