v1

type CleanPodPolicy string

type ElasticPolicy struct {
	// minReplicas is the lower limit for the number of replicas to which the training job
	// can scale down.  It defaults to null.
	// +optional
	MinReplicas *int32 `json:"minReplicas,omitempty"`
	// upper limit for the number of pods that can be set by the autoscaler; cannot be smaller than MinReplicas, defaults to null.
	// +optional
	MaxReplicas *int32 `json:"maxReplicas,omitempty"`

	RDZVBackend *RDZVBackend `json:"rdzvBackend,omitempty"`
	RDZVPort    *int32       `json:"rdzvPort,omitempty"`
	RDZVHost    *string      `json:"rdzvHost,omitempty"`
	RDZVID      *string      `json:"rdzvId,omitempty"`
	// RDZVConf contains additional rendezvous configuration (<key1>=<value1>,<key2>=<value2>,...).
	RDZVConf []RDZVConf `json:"rdzvConf,omitempty"`
	// Start a local standalone rendezvous backend that is represented by a C10d TCP store
	// on port 29400. Useful when launching single-node, multi-worker job. If specified
	// --rdzv_backend, --rdzv_endpoint, --rdzv_id are auto-assigned; any explicitly set values
	// are ignored.
	Standalone *bool `json:"standalone,omitempty"`
	// Number of workers per node; supported values: [auto, cpu, gpu, int].
	// Deprecated: This API is deprecated in v1.7+
	// Use .spec.nprocPerNode instead.
	NProcPerNode *int32 `json:"nProcPerNode,omitempty"`

	MaxRestarts *int32 `json:"maxRestarts,omitempty"`

	// Metrics contains the specifications which are used to calculate the
	// desired replica count (the maximum replica count across all metrics will
	// be used).  The desired replica count is calculated with multiplying the
	// ratio between the target value and the current value by the current
	// number of pods. Ergo, metrics used must decrease as the pod count is
	// increased, and vice-versa.  See the individual metric source types for
	// more information about how each type of metric must respond.
	// If not set, the HPA will not be created.
	// +optional
	Metrics []autoscalingv2.MetricSpec `json:"metrics,omitempty"`
}

type JobCondition struct {
	// Type of job condition.
	Type JobConditionType `json:"type"`
	// Status of the condition, one of True, False, Unknown.
	Status v1.ConditionStatus `json:"status"`
	// The reason for the condition's last transition.
	Reason string `json:"reason,omitempty"`
	// A human readable message indicating details about the transition.
	Message string `json:"message,omitempty"`
	// The last time this condition was updated.
	LastUpdateTime metav1.Time `json:"lastUpdateTime,omitempty"`
	// Last time the condition transitioned from one status to another.
	LastTransitionTime metav1.Time `json:"lastTransitionTime,omitempty"`
}

type JobConditionType string

type JobModeType string

type JobStatus struct {
	// Conditions is an array of current observed job conditions.
	Conditions []JobCondition `json:"conditions,omitempty"`

	// ReplicaStatuses is map of ReplicaType and ReplicaStatus,
	// specifies the status of each replica.
	ReplicaStatuses map[ReplicaType]*ReplicaStatus `json:"replicaStatuses,omitempty"`

	// Represents time when the job was acknowledged by the job controller.
	// It is not guaranteed to be set in happens-before order across separate operations.
	// It is represented in RFC3339 form and is in UTC.
	StartTime *metav1.Time `json:"startTime,omitempty"`

	// Represents time when the job was completed. It is not guaranteed to
	// be set in happens-before order across separate operations.
	// It is represented in RFC3339 form and is in UTC.
	CompletionTime *metav1.Time `json:"completionTime,omitempty"`

	// Represents last time when the job was reconciled. It is not guaranteed to
	// be set in happens-before order across separate operations.
	// It is represented in RFC3339 form and is in UTC.
	LastReconcileTime *metav1.Time `json:"lastReconcileTime,omitempty"`
}

type MPIJob struct {
	metav1.TypeMeta   `json:",inline"`
	metav1.ObjectMeta `json:"metadata,omitempty"`
	Spec              MPIJobSpec `json:"spec,omitempty"`
	Status            JobStatus  `json:"status,omitempty"`
}

type MPIJobList struct {
	metav1.TypeMeta `json:",inline"`
	metav1.ListMeta `json:"metadata,omitempty"`
	Items           []MPIJob `json:"items"`
}

type MPIJobSpec struct {

	// Specifies the number of slots per worker used in hostfile.
	// Defaults to 1.
	// +optional
	SlotsPerWorker *int32 `json:"slotsPerWorker,omitempty"`

	// CleanPodPolicy defines the policy that whether to kill pods after the job completes.
	// Defaults to None.
	CleanPodPolicy *CleanPodPolicy `json:"cleanPodPolicy,omitempty"`

	// `MPIReplicaSpecs` contains maps from `MPIReplicaType` to `ReplicaSpec` that
	// specify the MPI replicas to run.
	MPIReplicaSpecs map[ReplicaType]*ReplicaSpec `json:"mpiReplicaSpecs"`

	// MainContainer specifies name of the main container which
	// executes the MPI code.
	MainContainer string `json:"mainContainer,omitempty"`

	// `RunPolicy` encapsulates various runtime policies of the distributed training
	// job, for example how to clean up resources and how long the job can stay
	// active.
	RunPolicy RunPolicy `json:"runPolicy,omitempty"`
}

type MXJob struct {
	metav1.TypeMeta   `json:",inline"`
	metav1.ObjectMeta `json:"metadata,omitempty"`

	Spec   MXJobSpec `json:"spec,omitempty"`
	Status JobStatus `json:"status,omitempty"`
}

type MXJobList struct {
	metav1.TypeMeta `json:",inline"`
	metav1.ListMeta `json:"metadata,omitempty"`
	Items           []MXJob `json:"items"`
}

type MXJobSpec struct {
	// RunPolicy encapsulates various runtime policies of the distributed training
	// job, for example how to clean up resources and how long the job can stay
	// active.
	//+kubebuilder:validation:Optional
	RunPolicy RunPolicy `json:"runPolicy"`

	// JobMode specify the kind of MXjob to do. Different mode may have
	// different MXReplicaSpecs request
	JobMode JobModeType `json:"jobMode"`

	// MXReplicaSpecs is map of ReplicaType and ReplicaSpec
	// specifies the MX replicas to run.
	// For example,
	//   {
	//     "Scheduler": ReplicaSpec,
	//     "Server": ReplicaSpec,
	//     "Worker": ReplicaSpec,
	//   }
	MXReplicaSpecs map[ReplicaType]*ReplicaSpec `json:"mxReplicaSpecs"`
}

type MXJobStatus struct {
}

type PaddleElasticPolicy struct {
	// minReplicas is the lower limit for the number of replicas to which the training job
	// can scale down.  It defaults to null.
	// +optional
	MinReplicas *int32 `json:"minReplicas,omitempty"`
	// upper limit for the number of pods that can be set by the autoscaler; cannot be smaller than MinReplicas, defaults to null.
	// +optional
	MaxReplicas *int32 `json:"maxReplicas,omitempty"`

	// MaxRestarts is the limit for restart times of pods in elastic mode.
	// +optional
	MaxRestarts *int32 `json:"maxRestarts,omitempty"`

	// Metrics contains the specifications which are used to calculate the
	// desired replica count (the maximum replica count across all metrics will
	// be used).  The desired replica count is calculated with multiplying the
	// ratio between the target value and the current value by the current
	// number of pods. Ergo, metrics used must decrease as the pod count is
	// increased, and vice-versa.  See the individual metric source types for
	// more information about how each type of metric must respond.
	// If not set, the HPA will not be created.
	// +optional
	Metrics []autoscalingv2.MetricSpec `json:"metrics,omitempty"`
}

type PaddleJob struct {
	// Standard Kubernetes type metadata.
	metav1.TypeMeta `json:",inline"`

	metav1.ObjectMeta `json:"metadata,omitempty"`

	// Specification of the desired state of the PaddleJob.
	Spec PaddleJobSpec `json:"spec,omitempty"`

	// Most recently observed status of the PaddleJob.
	// Read-only (modified by the system).
	Status JobStatus `json:"status,omitempty"`
}

type PaddleJobList struct {
	// Standard type metadata.
	metav1.TypeMeta `json:",inline"`

	// Standard list metadata.
	metav1.ListMeta `json:"metadata,omitempty"`

	// List of PaddleJobs.
	Items []PaddleJob `json:"items"`
}

type PaddleJobSpec struct {
	// RunPolicy encapsulates various runtime policies of the distributed training
	// job, for example how to clean up resources and how long the job can stay
	// active.
	//+kubebuilder:validation:Optional
	RunPolicy RunPolicy `json:"runPolicy"`

	// ElasticPolicy holds the elastic policy for paddle job.
	ElasticPolicy *PaddleElasticPolicy `json:"elasticPolicy,omitempty"`

	// A map of PaddleReplicaType (type) to ReplicaSpec (value). Specifies the Paddle cluster configuration.
	// For example,
	//   {
	//     "Master": PaddleReplicaSpec,
	//     "Worker": PaddleReplicaSpec,
	//   }
	PaddleReplicaSpecs map[ReplicaType]*ReplicaSpec `json:"paddleReplicaSpecs"`
}

type PyTorchJob struct {
	// Standard Kubernetes type metadata.
	metav1.TypeMeta `json:",inline"`

	metav1.ObjectMeta `json:"metadata,omitempty"`

	// Specification of the desired state of the PyTorchJob.
	Spec PyTorchJobSpec `json:"spec,omitempty"`

	// Most recently observed status of the PyTorchJob.
	// Read-only (modified by the system).
	Status JobStatus `json:"status,omitempty"`
}

type PyTorchJobList struct {
	// Standard type metadata.
	metav1.TypeMeta `json:",inline"`

	// Standard list metadata.
	metav1.ListMeta `json:"metadata,omitempty"`

	// List of PyTorchJobs.
	Items []PyTorchJob `json:"items"`
}

type PyTorchJobSpec struct {
	// RunPolicy encapsulates various runtime policies of the distributed training
	// job, for example how to clean up resources and how long the job can stay
	// active.
	//+kubebuilder:validation:Optional
	RunPolicy RunPolicy `json:"runPolicy"`

	ElasticPolicy *ElasticPolicy `json:"elasticPolicy,omitempty"`

	// A map of PyTorchReplicaType (type) to ReplicaSpec (value). Specifies the PyTorch cluster configuration.
	// For example,
	//   {
	//     "Master": PyTorchReplicaSpec,
	//     "Worker": PyTorchReplicaSpec,
	//   }
	PyTorchReplicaSpecs map[ReplicaType]*ReplicaSpec `json:"pytorchReplicaSpecs"`

	// Number of workers per node; supported values: [auto, cpu, gpu, int].
	// For more, https://github.com/pytorch/pytorch/blob/26f7f470df64d90e092081e39507e4ac751f55d6/torch/distributed/run.py#L629-L658.
	// Defaults to auto.
	NprocPerNode *string `json:"nprocPerNode,omitempty"`
}

type RDZVBackend string

type RDZVConf struct {
	Key   string `json:"key,omitempty"`
	Value string `json:"value,omitempty"`
}

type ReplicaSpec struct {
	// Replicas is the desired number of replicas of the given template.
	// If unspecified, defaults to 1.
	Replicas *int32 `json:"replicas,omitempty"`

	// Template is the object that describes the pod that
	// will be created for this replica. RestartPolicy in PodTemplateSpec
	// will be overide by RestartPolicy in ReplicaSpec
	Template v1.PodTemplateSpec `json:"template,omitempty"`

	// Restart policy for all replicas within the job.
	// One of Always, OnFailure, Never and ExitCode.
	// Default to Never.
	RestartPolicy RestartPolicy `json:"restartPolicy,omitempty"`
}

type ReplicaStatus struct {
	// The number of actively running pods.
	Active int32 `json:"active,omitempty"`

	// The number of pods which reached phase Succeeded.
	Succeeded int32 `json:"succeeded,omitempty"`

	// The number of pods which reached phase Failed.
	Failed int32 `json:"failed,omitempty"`

	// Deprecated: Use Selector instead
	LabelSelector *metav1.LabelSelector `json:"labelSelector,omitempty"`

	// A Selector is a label query over a set of resources. The result of matchLabels and
	// matchExpressions are ANDed. An empty Selector matches all objects. A null
	// Selector matches no objects.
	Selector string `json:"selector,omitempty"`
}

type ReplicaType string

type RestartPolicy string

type RunPolicy struct {
	// CleanPodPolicy defines the policy to kill pods after the job completes.
	// Default to None.
	CleanPodPolicy *CleanPodPolicy `json:"cleanPodPolicy,omitempty"`

	// TTLSecondsAfterFinished is the TTL to clean up jobs.
	// It may take extra ReconcilePeriod seconds for the cleanup, since
	// reconcile gets called periodically.
	// Default to infinite.
	TTLSecondsAfterFinished *int32 `json:"ttlSecondsAfterFinished,omitempty"`

	// Specifies the duration in seconds relative to the startTime that the job may be active
	// before the system tries to terminate it; value must be positive integer.
	// +optional
	ActiveDeadlineSeconds *int64 `json:"activeDeadlineSeconds,omitempty"`

	// Optional number of retries before marking this job failed.
	// +optional
	BackoffLimit *int32 `json:"backoffLimit,omitempty"`

	// SchedulingPolicy defines the policy related to scheduling, e.g. gang-scheduling
	// +optional
	SchedulingPolicy *SchedulingPolicy `json:"schedulingPolicy,omitempty"`

	// suspend specifies whether the Job controller should create Pods or not.
	// If a Job is created with suspend set to true, no Pods are created by
	// the Job controller. If a Job is suspended after creation (i.e. the
	// flag goes from false to true), the Job controller will delete all
	// active Pods and PodGroups associated with this Job.
	// Users must design their workload to gracefully handle this.
	// Suspending a Job will reset the StartTime field of the Job.
	//
	// Defaults to false.
	// +kubebuilder:default:=false
	// +optional
	Suspend *bool `json:"suspend,omitempty"`
}

type SchedulingPolicy struct {
	MinAvailable           *int32                                 `json:"minAvailable,omitempty"`
	Queue                  string                                 `json:"queue,omitempty"`
	MinResources           *map[v1.ResourceName]resource.Quantity `json:"minResources,omitempty"`
	PriorityClass          string                                 `json:"priorityClass,omitempty"`
	ScheduleTimeoutSeconds *int32                                 `json:"scheduleTimeoutSeconds,omitempty"`
}

type SuccessPolicy string

type TFJob struct {
	// Standard Kubernetes type metadata.
	metav1.TypeMeta `json:",inline"`

	// +optional
	metav1.ObjectMeta `json:"metadata,omitempty"`

	// Specification of the desired state of the TFJob.
	// +optional
	Spec TFJobSpec `json:"spec,omitempty"`

	// Most recently observed status of the TFJob.
	// Populated by the system.
	// Read-only.
	// +optional
	Status JobStatus `json:"status,omitempty"`
}

type TFJobList struct {
	// Standard type metadata.
	metav1.TypeMeta `json:",inline"`

	// Standard list metadata.
	// +optional
	metav1.ListMeta `json:"metadata,omitempty"`

	// List of TFJobs.
	Items []TFJob `json:"items"`
}

type TFJobSpec struct {
	// RunPolicy encapsulates various runtime policies of the distributed training
	// job, for example how to clean up resources and how long the job can stay
	// active.
	//+kubebuilder:validation:Optional
	RunPolicy RunPolicy `json:"runPolicy"`

	// SuccessPolicy defines the policy to mark the TFJob as succeeded.
	// Default to "", using the default rules.
	// +optional
	SuccessPolicy *SuccessPolicy `json:"successPolicy,omitempty"`

	// A map of TFReplicaType (type) to ReplicaSpec (value). Specifies the TF cluster configuration.
	// For example,
	//   {
	//     "PS": ReplicaSpec,
	//     "Worker": ReplicaSpec,
	//   }
	TFReplicaSpecs map[ReplicaType]*ReplicaSpec `json:"tfReplicaSpecs"`

	// A switch to enable dynamic worker
	EnableDynamicWorker bool `json:"enableDynamicWorker,omitempty"`
}

type XGBoostJob struct {
	metav1.TypeMeta   `json:",inline"`
	metav1.ObjectMeta `json:"metadata,omitempty"`

	Spec   XGBoostJobSpec `json:"spec,omitempty"`
	Status JobStatus      `json:"status,omitempty"`
}

type XGBoostJobList struct {
	metav1.TypeMeta `json:",inline"`
	metav1.ListMeta `json:"metadata,omitempty"`
	Items           []XGBoostJob `json:"items"`
}

type XGBoostJobSpec struct {
	// INSERT ADDITIONAL SPEC FIELDS - desired state of cluster
	// Important: Run "make" to regenerate code after modifying this file
	//+kubebuilder:validation:Optional
	RunPolicy RunPolicy `json:"runPolicy"`

	XGBReplicaSpecs map[ReplicaType]*ReplicaSpec `json:"xgbReplicaSpecs"`
}