plugins

package
v1.5.21 Latest Latest
Warning

This package is not in the latest version of its module.

Go to latest
Published: Sep 27, 2023 License: Apache-2.0 Imports: 14 Imported by: 4

Documentation

Index

Constants

This section is empty.

Variables

View Source
var CleanPodPolicy_name = map[int32]string{
	0: "CLEANPOD_POLICY_NONE",
	1: "CLEANPOD_POLICY_RUNNING",
	2: "CLEANPOD_POLICY_ALL",
}
View Source
var CleanPodPolicy_value = map[string]int32{
	"CLEANPOD_POLICY_NONE":    0,
	"CLEANPOD_POLICY_RUNNING": 1,
	"CLEANPOD_POLICY_ALL":     2,
}
View Source
var RestartPolicy_name = map[int32]string{
	0: "RESTART_POLICY_NEVER",
	1: "RESTART_POLICY_ON_FAILURE",
	2: "RESTART_POLICY_ALWAYS",
}
View Source
var RestartPolicy_value = map[string]int32{
	"RESTART_POLICY_NEVER":      0,
	"RESTART_POLICY_ON_FAILURE": 1,
	"RESTART_POLICY_ALWAYS":     2,
}

Functions

This section is empty.

Types

type CleanPodPolicy

type CleanPodPolicy int32
const (
	CleanPodPolicy_CLEANPOD_POLICY_NONE    CleanPodPolicy = 0
	CleanPodPolicy_CLEANPOD_POLICY_RUNNING CleanPodPolicy = 1
	CleanPodPolicy_CLEANPOD_POLICY_ALL     CleanPodPolicy = 2
)

func (CleanPodPolicy) EnumDescriptor

func (CleanPodPolicy) EnumDescriptor() ([]byte, []int)

func (CleanPodPolicy) String

func (x CleanPodPolicy) String() string

type DistributedMPITrainingReplicaSpec

type DistributedMPITrainingReplicaSpec struct {
	// Number of replicas
	Replicas int32 `protobuf:"varint,1,opt,name=replicas,proto3" json:"replicas,omitempty"`
	// Image used for the replica group
	Image string `protobuf:"bytes,2,opt,name=image,proto3" json:"image,omitempty"`
	// Resources required for the replica group
	Resources *core.Resources `protobuf:"bytes,3,opt,name=resources,proto3" json:"resources,omitempty"`
	// Restart policy determines whether pods will be restarted when they exit
	RestartPolicy RestartPolicy `` /* 146-byte string literal not displayed */
	// MPI sometimes requires different command set for different replica groups
	Command              []string `protobuf:"bytes,5,rep,name=command,proto3" json:"command,omitempty"`
	XXX_NoUnkeyedLiteral struct{} `json:"-"`
	XXX_unrecognized     []byte   `json:"-"`
	XXX_sizecache        int32    `json:"-"`
}

Replica specification for distributed MPI training

func (*DistributedMPITrainingReplicaSpec) Descriptor

func (*DistributedMPITrainingReplicaSpec) Descriptor() ([]byte, []int)

func (*DistributedMPITrainingReplicaSpec) GetCommand

func (m *DistributedMPITrainingReplicaSpec) GetCommand() []string

func (*DistributedMPITrainingReplicaSpec) GetImage

func (*DistributedMPITrainingReplicaSpec) GetReplicas

func (m *DistributedMPITrainingReplicaSpec) GetReplicas() int32

func (*DistributedMPITrainingReplicaSpec) GetResources

func (*DistributedMPITrainingReplicaSpec) GetRestartPolicy

func (m *DistributedMPITrainingReplicaSpec) GetRestartPolicy() RestartPolicy

func (*DistributedMPITrainingReplicaSpec) ProtoMessage

func (*DistributedMPITrainingReplicaSpec) ProtoMessage()

func (*DistributedMPITrainingReplicaSpec) Reset

func (*DistributedMPITrainingReplicaSpec) String

func (*DistributedMPITrainingReplicaSpec) Validate

Validate checks the field values on DistributedMPITrainingReplicaSpec with the rules defined in the proto definition for this message. If any rules are violated, an error is returned.

func (*DistributedMPITrainingReplicaSpec) XXX_DiscardUnknown

func (m *DistributedMPITrainingReplicaSpec) XXX_DiscardUnknown()

func (*DistributedMPITrainingReplicaSpec) XXX_Marshal

func (m *DistributedMPITrainingReplicaSpec) XXX_Marshal(b []byte, deterministic bool) ([]byte, error)

func (*DistributedMPITrainingReplicaSpec) XXX_Merge

func (*DistributedMPITrainingReplicaSpec) XXX_Size

func (m *DistributedMPITrainingReplicaSpec) XXX_Size() int

func (*DistributedMPITrainingReplicaSpec) XXX_Unmarshal

func (m *DistributedMPITrainingReplicaSpec) XXX_Unmarshal(b []byte) error

type DistributedMPITrainingReplicaSpecValidationError

type DistributedMPITrainingReplicaSpecValidationError struct {
	// contains filtered or unexported fields
}

DistributedMPITrainingReplicaSpecValidationError is the validation error returned by DistributedMPITrainingReplicaSpec.Validate if the designated constraints aren't met.

func (DistributedMPITrainingReplicaSpecValidationError) Cause

Cause function returns cause value.

func (DistributedMPITrainingReplicaSpecValidationError) Error

Error satisfies the builtin error interface

func (DistributedMPITrainingReplicaSpecValidationError) ErrorName

ErrorName returns error name.

func (DistributedMPITrainingReplicaSpecValidationError) Field

Field function returns field value.

func (DistributedMPITrainingReplicaSpecValidationError) Key

Key function returns key value.

func (DistributedMPITrainingReplicaSpecValidationError) Reason

Reason function returns reason value.

type DistributedMPITrainingTask

type DistributedMPITrainingTask struct {
	// Worker replicas spec
	WorkerReplicas *DistributedMPITrainingReplicaSpec `protobuf:"bytes,1,opt,name=worker_replicas,json=workerReplicas,proto3" json:"worker_replicas,omitempty"`
	// Master replicas spec
	LauncherReplicas *DistributedMPITrainingReplicaSpec `protobuf:"bytes,2,opt,name=launcher_replicas,json=launcherReplicas,proto3" json:"launcher_replicas,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 `protobuf:"bytes,3,opt,name=run_policy,json=runPolicy,proto3" json:"run_policy,omitempty"`
	// Number of slots per worker
	Slots                int32    `protobuf:"varint,4,opt,name=slots,proto3" json:"slots,omitempty"`
	XXX_NoUnkeyedLiteral struct{} `json:"-"`
	XXX_unrecognized     []byte   `json:"-"`
	XXX_sizecache        int32    `json:"-"`
}

Proto for plugin that enables distributed training using https://github.com/kubeflow/mpi-operator

func (*DistributedMPITrainingTask) Descriptor

func (*DistributedMPITrainingTask) Descriptor() ([]byte, []int)

func (*DistributedMPITrainingTask) GetLauncherReplicas

func (*DistributedMPITrainingTask) GetRunPolicy

func (m *DistributedMPITrainingTask) GetRunPolicy() *RunPolicy

func (*DistributedMPITrainingTask) GetSlots

func (m *DistributedMPITrainingTask) GetSlots() int32

func (*DistributedMPITrainingTask) GetWorkerReplicas

func (*DistributedMPITrainingTask) ProtoMessage

func (*DistributedMPITrainingTask) ProtoMessage()

func (*DistributedMPITrainingTask) Reset

func (m *DistributedMPITrainingTask) Reset()

func (*DistributedMPITrainingTask) String

func (m *DistributedMPITrainingTask) String() string

func (*DistributedMPITrainingTask) Validate

func (m *DistributedMPITrainingTask) Validate() error

Validate checks the field values on DistributedMPITrainingTask with the rules defined in the proto definition for this message. If any rules are violated, an error is returned.

func (*DistributedMPITrainingTask) XXX_DiscardUnknown

func (m *DistributedMPITrainingTask) XXX_DiscardUnknown()

func (*DistributedMPITrainingTask) XXX_Marshal

func (m *DistributedMPITrainingTask) XXX_Marshal(b []byte, deterministic bool) ([]byte, error)

func (*DistributedMPITrainingTask) XXX_Merge

func (m *DistributedMPITrainingTask) XXX_Merge(src proto.Message)

func (*DistributedMPITrainingTask) XXX_Size

func (m *DistributedMPITrainingTask) XXX_Size() int

func (*DistributedMPITrainingTask) XXX_Unmarshal

func (m *DistributedMPITrainingTask) XXX_Unmarshal(b []byte) error

type DistributedMPITrainingTaskValidationError

type DistributedMPITrainingTaskValidationError struct {
	// contains filtered or unexported fields
}

DistributedMPITrainingTaskValidationError is the validation error returned by DistributedMPITrainingTask.Validate if the designated constraints aren't met.

func (DistributedMPITrainingTaskValidationError) Cause

Cause function returns cause value.

func (DistributedMPITrainingTaskValidationError) Error

Error satisfies the builtin error interface

func (DistributedMPITrainingTaskValidationError) ErrorName

ErrorName returns error name.

func (DistributedMPITrainingTaskValidationError) Field

Field function returns field value.

func (DistributedMPITrainingTaskValidationError) Key

Key function returns key value.

func (DistributedMPITrainingTaskValidationError) Reason

Reason function returns reason value.

type DistributedPyTorchTrainingReplicaSpec

type DistributedPyTorchTrainingReplicaSpec struct {
	// Number of replicas
	Replicas int32 `protobuf:"varint,1,opt,name=replicas,proto3" json:"replicas,omitempty"`
	// Image used for the replica group
	Image string `protobuf:"bytes,2,opt,name=image,proto3" json:"image,omitempty"`
	// Resources required for the replica group
	Resources *core.Resources `protobuf:"bytes,3,opt,name=resources,proto3" json:"resources,omitempty"`
	// RestartPolicy determines whether pods will be restarted when they exit
	RestartPolicy        RestartPolicy `` /* 146-byte string literal not displayed */
	XXX_NoUnkeyedLiteral struct{}      `json:"-"`
	XXX_unrecognized     []byte        `json:"-"`
	XXX_sizecache        int32         `json:"-"`
}

func (*DistributedPyTorchTrainingReplicaSpec) Descriptor

func (*DistributedPyTorchTrainingReplicaSpec) Descriptor() ([]byte, []int)

func (*DistributedPyTorchTrainingReplicaSpec) GetImage

func (*DistributedPyTorchTrainingReplicaSpec) GetReplicas

func (*DistributedPyTorchTrainingReplicaSpec) GetResources

func (*DistributedPyTorchTrainingReplicaSpec) GetRestartPolicy

func (*DistributedPyTorchTrainingReplicaSpec) ProtoMessage

func (*DistributedPyTorchTrainingReplicaSpec) ProtoMessage()

func (*DistributedPyTorchTrainingReplicaSpec) Reset

func (*DistributedPyTorchTrainingReplicaSpec) String

func (*DistributedPyTorchTrainingReplicaSpec) Validate

Validate checks the field values on DistributedPyTorchTrainingReplicaSpec with the rules defined in the proto definition for this message. If any rules are violated, an error is returned.

func (*DistributedPyTorchTrainingReplicaSpec) XXX_DiscardUnknown

func (m *DistributedPyTorchTrainingReplicaSpec) XXX_DiscardUnknown()

func (*DistributedPyTorchTrainingReplicaSpec) XXX_Marshal

func (m *DistributedPyTorchTrainingReplicaSpec) XXX_Marshal(b []byte, deterministic bool) ([]byte, error)

func (*DistributedPyTorchTrainingReplicaSpec) XXX_Merge

func (*DistributedPyTorchTrainingReplicaSpec) XXX_Size

func (*DistributedPyTorchTrainingReplicaSpec) XXX_Unmarshal

func (m *DistributedPyTorchTrainingReplicaSpec) XXX_Unmarshal(b []byte) error

type DistributedPyTorchTrainingReplicaSpecValidationError

type DistributedPyTorchTrainingReplicaSpecValidationError struct {
	// contains filtered or unexported fields
}

DistributedPyTorchTrainingReplicaSpecValidationError is the validation error returned by DistributedPyTorchTrainingReplicaSpec.Validate if the designated constraints aren't met.

func (DistributedPyTorchTrainingReplicaSpecValidationError) Cause

Cause function returns cause value.

func (DistributedPyTorchTrainingReplicaSpecValidationError) Error

Error satisfies the builtin error interface

func (DistributedPyTorchTrainingReplicaSpecValidationError) ErrorName

ErrorName returns error name.

func (DistributedPyTorchTrainingReplicaSpecValidationError) Field

Field function returns field value.

func (DistributedPyTorchTrainingReplicaSpecValidationError) Key

Key function returns key value.

func (DistributedPyTorchTrainingReplicaSpecValidationError) Reason

Reason function returns reason value.

type DistributedPyTorchTrainingTask

type DistributedPyTorchTrainingTask struct {
	// Worker replicas spec
	WorkerReplicas *DistributedPyTorchTrainingReplicaSpec `protobuf:"bytes,1,opt,name=worker_replicas,json=workerReplicas,proto3" json:"worker_replicas,omitempty"`
	// Master replicas spec, master replicas can only have 1 replica
	MasterReplicas *DistributedPyTorchTrainingReplicaSpec `protobuf:"bytes,2,opt,name=master_replicas,json=masterReplicas,proto3" json:"master_replicas,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 `protobuf:"bytes,3,opt,name=run_policy,json=runPolicy,proto3" json:"run_policy,omitempty"`
	// config for an elastic pytorch job
	ElasticConfig        *ElasticConfig `protobuf:"bytes,4,opt,name=elastic_config,json=elasticConfig,proto3" json:"elastic_config,omitempty"`
	XXX_NoUnkeyedLiteral struct{}       `json:"-"`
	XXX_unrecognized     []byte         `json:"-"`
	XXX_sizecache        int32          `json:"-"`
}

Proto for plugin that enables distributed training using https://github.com/kubeflow/pytorch-operator

func (*DistributedPyTorchTrainingTask) Descriptor

func (*DistributedPyTorchTrainingTask) Descriptor() ([]byte, []int)

func (*DistributedPyTorchTrainingTask) GetElasticConfig

func (m *DistributedPyTorchTrainingTask) GetElasticConfig() *ElasticConfig

func (*DistributedPyTorchTrainingTask) GetMasterReplicas

func (*DistributedPyTorchTrainingTask) GetRunPolicy

func (m *DistributedPyTorchTrainingTask) GetRunPolicy() *RunPolicy

func (*DistributedPyTorchTrainingTask) GetWorkerReplicas

func (*DistributedPyTorchTrainingTask) ProtoMessage

func (*DistributedPyTorchTrainingTask) ProtoMessage()

func (*DistributedPyTorchTrainingTask) Reset

func (m *DistributedPyTorchTrainingTask) Reset()

func (*DistributedPyTorchTrainingTask) String

func (*DistributedPyTorchTrainingTask) Validate

func (m *DistributedPyTorchTrainingTask) Validate() error

Validate checks the field values on DistributedPyTorchTrainingTask with the rules defined in the proto definition for this message. If any rules are violated, an error is returned.

func (*DistributedPyTorchTrainingTask) XXX_DiscardUnknown

func (m *DistributedPyTorchTrainingTask) XXX_DiscardUnknown()

func (*DistributedPyTorchTrainingTask) XXX_Marshal

func (m *DistributedPyTorchTrainingTask) XXX_Marshal(b []byte, deterministic bool) ([]byte, error)

func (*DistributedPyTorchTrainingTask) XXX_Merge

func (m *DistributedPyTorchTrainingTask) XXX_Merge(src proto.Message)

func (*DistributedPyTorchTrainingTask) XXX_Size

func (m *DistributedPyTorchTrainingTask) XXX_Size() int

func (*DistributedPyTorchTrainingTask) XXX_Unmarshal

func (m *DistributedPyTorchTrainingTask) XXX_Unmarshal(b []byte) error

type DistributedPyTorchTrainingTaskValidationError

type DistributedPyTorchTrainingTaskValidationError struct {
	// contains filtered or unexported fields
}

DistributedPyTorchTrainingTaskValidationError is the validation error returned by DistributedPyTorchTrainingTask.Validate if the designated constraints aren't met.

func (DistributedPyTorchTrainingTaskValidationError) Cause

Cause function returns cause value.

func (DistributedPyTorchTrainingTaskValidationError) Error

Error satisfies the builtin error interface

func (DistributedPyTorchTrainingTaskValidationError) ErrorName

ErrorName returns error name.

func (DistributedPyTorchTrainingTaskValidationError) Field

Field function returns field value.

func (DistributedPyTorchTrainingTaskValidationError) Key

Key function returns key value.

func (DistributedPyTorchTrainingTaskValidationError) Reason

Reason function returns reason value.

type DistributedTensorflowTrainingReplicaSpec

type DistributedTensorflowTrainingReplicaSpec struct {
	// Number of replicas
	Replicas int32 `protobuf:"varint,1,opt,name=replicas,proto3" json:"replicas,omitempty"`
	// Image used for the replica group
	Image string `protobuf:"bytes,2,opt,name=image,proto3" json:"image,omitempty"`
	// Resources required for the replica group
	Resources *core.Resources `protobuf:"bytes,3,opt,name=resources,proto3" json:"resources,omitempty"`
	// RestartPolicy Determines whether pods will be restarted when they exit
	RestartPolicy        RestartPolicy `` /* 146-byte string literal not displayed */
	XXX_NoUnkeyedLiteral struct{}      `json:"-"`
	XXX_unrecognized     []byte        `json:"-"`
	XXX_sizecache        int32         `json:"-"`
}

func (*DistributedTensorflowTrainingReplicaSpec) Descriptor

func (*DistributedTensorflowTrainingReplicaSpec) Descriptor() ([]byte, []int)

func (*DistributedTensorflowTrainingReplicaSpec) GetImage

func (*DistributedTensorflowTrainingReplicaSpec) GetReplicas

func (*DistributedTensorflowTrainingReplicaSpec) GetResources

func (*DistributedTensorflowTrainingReplicaSpec) GetRestartPolicy

func (*DistributedTensorflowTrainingReplicaSpec) ProtoMessage

func (*DistributedTensorflowTrainingReplicaSpec) Reset

func (*DistributedTensorflowTrainingReplicaSpec) String

func (*DistributedTensorflowTrainingReplicaSpec) Validate

Validate checks the field values on DistributedTensorflowTrainingReplicaSpec with the rules defined in the proto definition for this message. If any rules are violated, an error is returned.

func (*DistributedTensorflowTrainingReplicaSpec) XXX_DiscardUnknown

func (m *DistributedTensorflowTrainingReplicaSpec) XXX_DiscardUnknown()

func (*DistributedTensorflowTrainingReplicaSpec) XXX_Marshal

func (m *DistributedTensorflowTrainingReplicaSpec) XXX_Marshal(b []byte, deterministic bool) ([]byte, error)

func (*DistributedTensorflowTrainingReplicaSpec) XXX_Merge

func (*DistributedTensorflowTrainingReplicaSpec) XXX_Size

func (*DistributedTensorflowTrainingReplicaSpec) XXX_Unmarshal

func (m *DistributedTensorflowTrainingReplicaSpec) XXX_Unmarshal(b []byte) error

type DistributedTensorflowTrainingReplicaSpecValidationError

type DistributedTensorflowTrainingReplicaSpecValidationError struct {
	// contains filtered or unexported fields
}

DistributedTensorflowTrainingReplicaSpecValidationError is the validation error returned by DistributedTensorflowTrainingReplicaSpec.Validate if the designated constraints aren't met.

func (DistributedTensorflowTrainingReplicaSpecValidationError) Cause

Cause function returns cause value.

func (DistributedTensorflowTrainingReplicaSpecValidationError) Error

Error satisfies the builtin error interface

func (DistributedTensorflowTrainingReplicaSpecValidationError) ErrorName

ErrorName returns error name.

func (DistributedTensorflowTrainingReplicaSpecValidationError) Field

Field function returns field value.

func (DistributedTensorflowTrainingReplicaSpecValidationError) Key

Key function returns key value.

func (DistributedTensorflowTrainingReplicaSpecValidationError) Reason

Reason function returns reason value.

type DistributedTensorflowTrainingTask

type DistributedTensorflowTrainingTask struct {
	// Worker replicas spec
	WorkerReplicas *DistributedTensorflowTrainingReplicaSpec `protobuf:"bytes,1,opt,name=worker_replicas,json=workerReplicas,proto3" json:"worker_replicas,omitempty"`
	// Parameter server replicas spec
	PsReplicas *DistributedTensorflowTrainingReplicaSpec `protobuf:"bytes,2,opt,name=ps_replicas,json=psReplicas,proto3" json:"ps_replicas,omitempty"`
	// Chief replicas spec
	ChiefReplicas *DistributedTensorflowTrainingReplicaSpec `protobuf:"bytes,3,opt,name=chief_replicas,json=chiefReplicas,proto3" json:"chief_replicas,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 `protobuf:"bytes,4,opt,name=run_policy,json=runPolicy,proto3" json:"run_policy,omitempty"`
	XXX_NoUnkeyedLiteral struct{}   `json:"-"`
	XXX_unrecognized     []byte     `json:"-"`
	XXX_sizecache        int32      `json:"-"`
}

Proto for plugin that enables distributed training using https://github.com/kubeflow/tf-operator

func (*DistributedTensorflowTrainingTask) Descriptor

func (*DistributedTensorflowTrainingTask) Descriptor() ([]byte, []int)

func (*DistributedTensorflowTrainingTask) GetChiefReplicas

func (*DistributedTensorflowTrainingTask) GetPsReplicas

func (*DistributedTensorflowTrainingTask) GetRunPolicy

func (m *DistributedTensorflowTrainingTask) GetRunPolicy() *RunPolicy

func (*DistributedTensorflowTrainingTask) GetWorkerReplicas

func (*DistributedTensorflowTrainingTask) ProtoMessage

func (*DistributedTensorflowTrainingTask) ProtoMessage()

func (*DistributedTensorflowTrainingTask) Reset

func (*DistributedTensorflowTrainingTask) String

func (*DistributedTensorflowTrainingTask) Validate

Validate checks the field values on DistributedTensorflowTrainingTask with the rules defined in the proto definition for this message. If any rules are violated, an error is returned.

func (*DistributedTensorflowTrainingTask) XXX_DiscardUnknown

func (m *DistributedTensorflowTrainingTask) XXX_DiscardUnknown()

func (*DistributedTensorflowTrainingTask) XXX_Marshal

func (m *DistributedTensorflowTrainingTask) XXX_Marshal(b []byte, deterministic bool) ([]byte, error)

func (*DistributedTensorflowTrainingTask) XXX_Merge

func (*DistributedTensorflowTrainingTask) XXX_Size

func (m *DistributedTensorflowTrainingTask) XXX_Size() int

func (*DistributedTensorflowTrainingTask) XXX_Unmarshal

func (m *DistributedTensorflowTrainingTask) XXX_Unmarshal(b []byte) error

type DistributedTensorflowTrainingTaskValidationError

type DistributedTensorflowTrainingTaskValidationError struct {
	// contains filtered or unexported fields
}

DistributedTensorflowTrainingTaskValidationError is the validation error returned by DistributedTensorflowTrainingTask.Validate if the designated constraints aren't met.

func (DistributedTensorflowTrainingTaskValidationError) Cause

Cause function returns cause value.

func (DistributedTensorflowTrainingTaskValidationError) Error

Error satisfies the builtin error interface

func (DistributedTensorflowTrainingTaskValidationError) ErrorName

ErrorName returns error name.

func (DistributedTensorflowTrainingTaskValidationError) Field

Field function returns field value.

func (DistributedTensorflowTrainingTaskValidationError) Key

Key function returns key value.

func (DistributedTensorflowTrainingTaskValidationError) Reason

Reason function returns reason value.

type ElasticConfig

type ElasticConfig struct {
	RdzvBackend          string   `protobuf:"bytes,1,opt,name=rdzv_backend,json=rdzvBackend,proto3" json:"rdzv_backend,omitempty"`
	MinReplicas          int32    `protobuf:"varint,2,opt,name=min_replicas,json=minReplicas,proto3" json:"min_replicas,omitempty"`
	MaxReplicas          int32    `protobuf:"varint,3,opt,name=max_replicas,json=maxReplicas,proto3" json:"max_replicas,omitempty"`
	NprocPerNode         int32    `protobuf:"varint,4,opt,name=nproc_per_node,json=nprocPerNode,proto3" json:"nproc_per_node,omitempty"`
	MaxRestarts          int32    `protobuf:"varint,5,opt,name=max_restarts,json=maxRestarts,proto3" json:"max_restarts,omitempty"`
	XXX_NoUnkeyedLiteral struct{} `json:"-"`
	XXX_unrecognized     []byte   `json:"-"`
	XXX_sizecache        int32    `json:"-"`
}

Custom proto for torch elastic config for distributed training using https://github.com/kubeflow/training-operator/blob/master/pkg/apis/kubeflow.org/v1/pytorch_types.go

func (*ElasticConfig) Descriptor

func (*ElasticConfig) Descriptor() ([]byte, []int)

func (*ElasticConfig) GetMaxReplicas

func (m *ElasticConfig) GetMaxReplicas() int32

func (*ElasticConfig) GetMaxRestarts

func (m *ElasticConfig) GetMaxRestarts() int32

func (*ElasticConfig) GetMinReplicas

func (m *ElasticConfig) GetMinReplicas() int32

func (*ElasticConfig) GetNprocPerNode

func (m *ElasticConfig) GetNprocPerNode() int32

func (*ElasticConfig) GetRdzvBackend

func (m *ElasticConfig) GetRdzvBackend() string

func (*ElasticConfig) ProtoMessage

func (*ElasticConfig) ProtoMessage()

func (*ElasticConfig) Reset

func (m *ElasticConfig) Reset()

func (*ElasticConfig) String

func (m *ElasticConfig) String() string

func (*ElasticConfig) Validate

func (m *ElasticConfig) Validate() error

Validate checks the field values on ElasticConfig with the rules defined in the proto definition for this message. If any rules are violated, an error is returned.

func (*ElasticConfig) XXX_DiscardUnknown

func (m *ElasticConfig) XXX_DiscardUnknown()

func (*ElasticConfig) XXX_Marshal

func (m *ElasticConfig) XXX_Marshal(b []byte, deterministic bool) ([]byte, error)

func (*ElasticConfig) XXX_Merge

func (m *ElasticConfig) XXX_Merge(src proto.Message)

func (*ElasticConfig) XXX_Size

func (m *ElasticConfig) XXX_Size() int

func (*ElasticConfig) XXX_Unmarshal

func (m *ElasticConfig) XXX_Unmarshal(b []byte) error

type ElasticConfigValidationError

type ElasticConfigValidationError struct {
	// contains filtered or unexported fields
}

ElasticConfigValidationError is the validation error returned by ElasticConfig.Validate if the designated constraints aren't met.

func (ElasticConfigValidationError) Cause

Cause function returns cause value.

func (ElasticConfigValidationError) Error

Error satisfies the builtin error interface

func (ElasticConfigValidationError) ErrorName

func (e ElasticConfigValidationError) ErrorName() string

ErrorName returns error name.

func (ElasticConfigValidationError) Field

Field function returns field value.

func (ElasticConfigValidationError) Key

Key function returns key value.

func (ElasticConfigValidationError) Reason

Reason function returns reason value.

type RestartPolicy

type RestartPolicy int32
const (
	RestartPolicy_RESTART_POLICY_NEVER      RestartPolicy = 0
	RestartPolicy_RESTART_POLICY_ON_FAILURE RestartPolicy = 1
	RestartPolicy_RESTART_POLICY_ALWAYS     RestartPolicy = 2
)

func (RestartPolicy) EnumDescriptor

func (RestartPolicy) EnumDescriptor() ([]byte, []int)

func (RestartPolicy) String

func (x RestartPolicy) String() string

type RunPolicy

type RunPolicy struct {
	// Defines the policy to kill pods after the job completes. Default to None.
	CleanPodPolicy CleanPodPolicy `` /* 152-byte string literal not displayed */
	// TTL to clean up jobs. Default to infinite.
	TtlSecondsAfterFinished int32 `` /* 135-byte string literal not displayed */
	// 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.
	ActiveDeadlineSeconds int32 `` /* 127-byte string literal not displayed */
	// Number of retries before marking this job failed.
	BackoffLimit         int32    `protobuf:"varint,4,opt,name=backoff_limit,json=backoffLimit,proto3" json:"backoff_limit,omitempty"`
	XXX_NoUnkeyedLiteral struct{} `json:"-"`
	XXX_unrecognized     []byte   `json:"-"`
	XXX_sizecache        int32    `json:"-"`
}

func (*RunPolicy) Descriptor

func (*RunPolicy) Descriptor() ([]byte, []int)

func (*RunPolicy) GetActiveDeadlineSeconds

func (m *RunPolicy) GetActiveDeadlineSeconds() int32

func (*RunPolicy) GetBackoffLimit

func (m *RunPolicy) GetBackoffLimit() int32

func (*RunPolicy) GetCleanPodPolicy

func (m *RunPolicy) GetCleanPodPolicy() CleanPodPolicy

func (*RunPolicy) GetTtlSecondsAfterFinished

func (m *RunPolicy) GetTtlSecondsAfterFinished() int32

func (*RunPolicy) ProtoMessage

func (*RunPolicy) ProtoMessage()

func (*RunPolicy) Reset

func (m *RunPolicy) Reset()

func (*RunPolicy) String

func (m *RunPolicy) String() string

func (*RunPolicy) Validate

func (m *RunPolicy) Validate() error

Validate checks the field values on RunPolicy with the rules defined in the proto definition for this message. If any rules are violated, an error is returned.

func (*RunPolicy) XXX_DiscardUnknown

func (m *RunPolicy) XXX_DiscardUnknown()

func (*RunPolicy) XXX_Marshal

func (m *RunPolicy) XXX_Marshal(b []byte, deterministic bool) ([]byte, error)

func (*RunPolicy) XXX_Merge

func (m *RunPolicy) XXX_Merge(src proto.Message)

func (*RunPolicy) XXX_Size

func (m *RunPolicy) XXX_Size() int

func (*RunPolicy) XXX_Unmarshal

func (m *RunPolicy) XXX_Unmarshal(b []byte) error

type RunPolicyValidationError

type RunPolicyValidationError struct {
	// contains filtered or unexported fields
}

RunPolicyValidationError is the validation error returned by RunPolicy.Validate if the designated constraints aren't met.

func (RunPolicyValidationError) Cause

func (e RunPolicyValidationError) Cause() error

Cause function returns cause value.

func (RunPolicyValidationError) Error

func (e RunPolicyValidationError) Error() string

Error satisfies the builtin error interface

func (RunPolicyValidationError) ErrorName

func (e RunPolicyValidationError) ErrorName() string

ErrorName returns error name.

func (RunPolicyValidationError) Field

func (e RunPolicyValidationError) Field() string

Field function returns field value.

func (RunPolicyValidationError) Key

Key function returns key value.

func (RunPolicyValidationError) Reason

func (e RunPolicyValidationError) Reason() string

Reason function returns reason value.

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL