K8s Custom Resource and Operator For TensorFlow jobs

Overview

TFJob provides a Kubernetes custom resource that makes it easy to run distributed or non-distributed TensorFlow jobs on Kubernetes.

For a complete reference of the custom resource definitions, please refer to the API Definition.
For details on its design, please refer to the v1alpha2 design doc.
For details on its obersibility, please refer to the monitoring design doc.

kubectl apply -f ./deploy/v1/tf-operator.yaml

Please refer to the quick-start-v1.md and Kubeflow user guide for more information.

You can:

This is a part of Kubeflow, so please see readme in kubeflow/kubeflow to get in touch with the community.

Please refer to the developer_guide

Please refer to CHANGELOG

The following table lists the most recent few versions of the operator.

Operator Version	API Version	Kubernetes Version
`latest` (master HEAD)	`v1`	1.16+
`v1.0.x`	`v1`	1.16+

Name		Name	Last commit message	Last commit date
Latest commit History 650 Commits
.github		.github
build/images/tf_operator		build/images/tf_operator
cmd/tf-operator.v1		cmd/tf-operator.v1
deploy/v1		deploy/v1
docs		docs
examples		examples
hack		hack
manifests		manifests
pkg		pkg
py/kubeflow		py/kubeflow
scripts		scripts
sdk/python		sdk/python
test		test
third_party/library		third_party/library
third_party_licenses		third_party_licenses
vendor		vendor
.gcloudignore		.gcloudignore
.gitignore		.gitignore
.pylintrc		.pylintrc
.style.yapf		.style.yapf
.travis.yml		.travis.yml
CHANGELOG.md		CHANGELOG.md
LICENSE		LICENSE
Makefile		Makefile
OWNERS		OWNERS
README.md		README.md
developer_guide.md		developer_guide.md
e2e_testing.md		e2e_testing.md
go.mod		go.mod
go.sum		go.sum
prow_config.yaml		prow_config.yaml
releasing.md		releasing.md
submit_release_job.sh		submit_release_job.sh
tf_job_design_doc.md		tf_job_design_doc.md
vendor.go		vendor.go