TFJob TensorFlow

Reference documentation for TFJob

Packages:

kubeflow.org

Package v1beta2 is the v1beta2 version of the API.

Resource Types:

TFJob

TFJob represents the configuration of signal TFJob

FieldDescription
apiVersionstringkubeflow.org/v1beta2
kindstringTFJob
metadataKubernetes meta/v1.ObjectMetaStandard object’s metadata.Refer to the Kubernetes API documentation for the fields of themetadata field.
specTFJobSpecSpecification of the desired behavior of the TFJob.
activeDeadlineSecondsint64(Optional)Specifies the duration in seconds relative to the startTime that the job may be activebefore the system tries to terminate it; value must be positive integer.This method applies only to pods with restartPolicy == OnFailure or Always.
backoffLimitint32(Optional)Optional number of retries before marking this job failed.
cleanPodPolicycommon/v1beta2.CleanPodPolicyCleanPodPolicy defines the policy to kill pods after TFJob issucceeded.Default to Running.
ttlSecondsAfterFinishedint32TTLSecondsAfterFinished is the TTL to clean up tf-jobs (temporarybefore kubernetes adds the cleanup controller).It may take extra ReconcilePeriod seconds for the cleanup, sincereconcile gets called periodically.Default to infinite.
tfReplicaSpecsmap[github.com/kubeflow/tf-operator/pkg/apis/tensorflow/v1beta2.TFReplicaType]*github.com/kubeflow/tf-operator/pkg/apis/common/v1beta2.ReplicaSpecTFReplicaSpecs is map of TFReplicaType and ReplicaSpecspecifies the TF replicas to run.For example,{“PS”: ReplicaSpec,“Worker”: ReplicaSpec,}

statuscommon/v1beta2.JobStatusMost recently observed status of the TFJob.This data may not be up to date.Populated by the system.Read-only.

TFJobSpec

(Appears on:TFJob)

TFJobSpec is a desired state description of the TFJob.

FieldDescription
activeDeadlineSecondsint64(Optional)Specifies the duration in seconds relative to the startTime that the job may be activebefore the system tries to terminate it; value must be positive integer.This method applies only to pods with restartPolicy == OnFailure or Always.
backoffLimitint32(Optional)Optional number of retries before marking this job failed.
cleanPodPolicycommon/v1beta2.CleanPodPolicyCleanPodPolicy defines the policy to kill pods after TFJob issucceeded.Default to Running.
ttlSecondsAfterFinishedint32TTLSecondsAfterFinished is the TTL to clean up tf-jobs (temporarybefore kubernetes adds the cleanup controller).It may take extra ReconcilePeriod seconds for the cleanup, sincereconcile gets called periodically.Default to infinite.
tfReplicaSpecsmap[github.com/kubeflow/tf-operator/pkg/apis/tensorflow/v1beta2.TFReplicaType]*github.com/kubeflow/tf-operator/pkg/apis/common/v1beta2.ReplicaSpecTFReplicaSpecs is map of TFReplicaType and ReplicaSpecspecifies the TF replicas to run.For example,{“PS”: ReplicaSpec,“Worker”: ReplicaSpec,}

TFReplicaType(string alias)

TFReplicaType is the type for TFReplica.


Generated with gen-crd-api-reference-docson git commit fd76deec.

Feedback

Was this page helpful?

Glad to hear it! Please tell us how we can improve.

Sorry to hear that. Please tell us how we can improve.

Last modified 17.06.2019: update tfjob, pytorchjob ref scripts, style tables (#805) (affc79c5)