Documentation ¶
Index ¶
- Constants
- Variables
- type CleanPodPolicy
- func (CleanPodPolicy) Descriptor() protoreflect.EnumDescriptor
- func (x CleanPodPolicy) Enum() *CleanPodPolicy
- func (CleanPodPolicy) EnumDescriptor() ([]byte, []int)deprecated
- func (x CleanPodPolicy) Number() protoreflect.EnumNumber
- func (x CleanPodPolicy) String() string
- func (CleanPodPolicy) Type() protoreflect.EnumType
- type CommonReplicaSpec
- type DistributedMPITrainingReplicaSpec
- func (*DistributedMPITrainingReplicaSpec) Descriptor() ([]byte, []int)deprecated
- func (x *DistributedMPITrainingReplicaSpec) GetCommand() []string
- func (x *DistributedMPITrainingReplicaSpec) GetCommon() *plugins.CommonReplicaSpec
- func (x *DistributedMPITrainingReplicaSpec) GetImage() stringdeprecated
- func (x *DistributedMPITrainingReplicaSpec) GetReplicas() int32deprecated
- func (x *DistributedMPITrainingReplicaSpec) GetResources() *core.Resourcesdeprecated
- func (x *DistributedMPITrainingReplicaSpec) GetRestartPolicy() plugins.RestartPolicydeprecated
- func (*DistributedMPITrainingReplicaSpec) ProtoMessage()
- func (x *DistributedMPITrainingReplicaSpec) ProtoReflect() protoreflect.Message
- func (x *DistributedMPITrainingReplicaSpec) Reset()
- func (x *DistributedMPITrainingReplicaSpec) String() string
- type DistributedMPITrainingTask
- func (*DistributedMPITrainingTask) Descriptor() ([]byte, []int)deprecated
- func (x *DistributedMPITrainingTask) GetLauncherReplicas() *DistributedMPITrainingReplicaSpec
- func (x *DistributedMPITrainingTask) GetRunPolicy() *RunPolicy
- func (x *DistributedMPITrainingTask) GetSlots() int32
- func (x *DistributedMPITrainingTask) GetWorkerReplicas() *DistributedMPITrainingReplicaSpec
- func (*DistributedMPITrainingTask) ProtoMessage()
- func (x *DistributedMPITrainingTask) ProtoReflect() protoreflect.Message
- func (x *DistributedMPITrainingTask) Reset()
- func (x *DistributedMPITrainingTask) String() string
- type DistributedPyTorchTrainingReplicaSpec
- func (*DistributedPyTorchTrainingReplicaSpec) Descriptor() ([]byte, []int)deprecated
- func (x *DistributedPyTorchTrainingReplicaSpec) GetCommon() *plugins.CommonReplicaSpec
- func (x *DistributedPyTorchTrainingReplicaSpec) GetImage() stringdeprecated
- func (x *DistributedPyTorchTrainingReplicaSpec) GetReplicas() int32deprecated
- func (x *DistributedPyTorchTrainingReplicaSpec) GetResources() *core.Resourcesdeprecated
- func (x *DistributedPyTorchTrainingReplicaSpec) GetRestartPolicy() plugins.RestartPolicydeprecated
- func (*DistributedPyTorchTrainingReplicaSpec) ProtoMessage()
- func (x *DistributedPyTorchTrainingReplicaSpec) ProtoReflect() protoreflect.Message
- func (x *DistributedPyTorchTrainingReplicaSpec) Reset()
- func (x *DistributedPyTorchTrainingReplicaSpec) String() string
- type DistributedPyTorchTrainingTask
- func (*DistributedPyTorchTrainingTask) Descriptor() ([]byte, []int)deprecated
- func (x *DistributedPyTorchTrainingTask) GetElasticConfig() *ElasticConfig
- func (x *DistributedPyTorchTrainingTask) GetMasterReplicas() *DistributedPyTorchTrainingReplicaSpec
- func (x *DistributedPyTorchTrainingTask) GetRunPolicy() *RunPolicy
- func (x *DistributedPyTorchTrainingTask) GetWorkerReplicas() *DistributedPyTorchTrainingReplicaSpec
- func (*DistributedPyTorchTrainingTask) ProtoMessage()
- func (x *DistributedPyTorchTrainingTask) ProtoReflect() protoreflect.Message
- func (x *DistributedPyTorchTrainingTask) Reset()
- func (x *DistributedPyTorchTrainingTask) String() string
- type DistributedTensorflowTrainingReplicaSpec
- func (*DistributedTensorflowTrainingReplicaSpec) Descriptor() ([]byte, []int)deprecated
- func (x *DistributedTensorflowTrainingReplicaSpec) GetCommon() *plugins.CommonReplicaSpec
- func (x *DistributedTensorflowTrainingReplicaSpec) GetImage() stringdeprecated
- func (x *DistributedTensorflowTrainingReplicaSpec) GetReplicas() int32deprecated
- func (x *DistributedTensorflowTrainingReplicaSpec) GetResources() *core.Resourcesdeprecated
- func (x *DistributedTensorflowTrainingReplicaSpec) GetRestartPolicy() plugins.RestartPolicydeprecated
- func (*DistributedTensorflowTrainingReplicaSpec) ProtoMessage()
- func (x *DistributedTensorflowTrainingReplicaSpec) ProtoReflect() protoreflect.Message
- func (x *DistributedTensorflowTrainingReplicaSpec) Reset()
- func (x *DistributedTensorflowTrainingReplicaSpec) String() string
- type DistributedTensorflowTrainingTask
- func (*DistributedTensorflowTrainingTask) Descriptor() ([]byte, []int)deprecated
- func (x *DistributedTensorflowTrainingTask) GetChiefReplicas() *DistributedTensorflowTrainingReplicaSpec
- func (x *DistributedTensorflowTrainingTask) GetEvaluatorReplicas() *DistributedTensorflowTrainingReplicaSpec
- func (x *DistributedTensorflowTrainingTask) GetPsReplicas() *DistributedTensorflowTrainingReplicaSpec
- func (x *DistributedTensorflowTrainingTask) GetRunPolicy() *RunPolicy
- func (x *DistributedTensorflowTrainingTask) GetWorkerReplicas() *DistributedTensorflowTrainingReplicaSpec
- func (*DistributedTensorflowTrainingTask) ProtoMessage()
- func (x *DistributedTensorflowTrainingTask) ProtoReflect() protoreflect.Message
- func (x *DistributedTensorflowTrainingTask) Reset()
- func (x *DistributedTensorflowTrainingTask) String() string
- type ElasticConfig
- func (*ElasticConfig) Descriptor() ([]byte, []int)deprecated
- func (x *ElasticConfig) GetMaxReplicas() int32
- func (x *ElasticConfig) GetMaxRestarts() int32
- func (x *ElasticConfig) GetMinReplicas() int32
- func (x *ElasticConfig) GetNprocPerNode() int32
- func (x *ElasticConfig) GetRdzvBackend() string
- func (*ElasticConfig) ProtoMessage()
- func (x *ElasticConfig) ProtoReflect() protoreflect.Message
- func (x *ElasticConfig) Reset()
- func (x *ElasticConfig) String() string
- type RestartPolicy
- type RunPolicy
- func (*RunPolicy) Descriptor() ([]byte, []int)deprecated
- func (x *RunPolicy) GetActiveDeadlineSeconds() int32
- func (x *RunPolicy) GetBackoffLimit() int32
- func (x *RunPolicy) GetCleanPodPolicy() CleanPodPolicy
- func (x *RunPolicy) GetTtlSecondsAfterFinished() int32
- func (*RunPolicy) ProtoMessage()
- func (x *RunPolicy) ProtoReflect() protoreflect.Message
- func (x *RunPolicy) Reset()
- func (x *RunPolicy) String() string
Constants ¶
const RestartPolicy_RESTART_POLICY_ALWAYS = plugins.RestartPolicy_RESTART_POLICY_ALWAYS
const RestartPolicy_RESTART_POLICY_NEVER = plugins.RestartPolicy_RESTART_POLICY_NEVER
const RestartPolicy_RESTART_POLICY_ON_FAILURE = plugins.RestartPolicy_RESTART_POLICY_ON_FAILURE
Variables ¶
var ( CleanPodPolicy_name = map[int32]string{ 0: "CLEANPOD_POLICY_NONE", 1: "CLEANPOD_POLICY_RUNNING", 2: "CLEANPOD_POLICY_ALL", } CleanPodPolicy_value = map[string]int32{ "CLEANPOD_POLICY_NONE": 0, "CLEANPOD_POLICY_RUNNING": 1, "CLEANPOD_POLICY_ALL": 2, } )
Enum value maps for CleanPodPolicy.
var File_flyteidl_plugins_kubeflow_common_proto protoreflect.FileDescriptor
var File_flyteidl_plugins_kubeflow_mpi_proto protoreflect.FileDescriptor
var File_flyteidl_plugins_kubeflow_pytorch_proto protoreflect.FileDescriptor
var File_flyteidl_plugins_kubeflow_tensorflow_proto protoreflect.FileDescriptor
var RestartPolicy_name = plugins.RestartPolicy_name
var RestartPolicy_value = plugins.RestartPolicy_value
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) Descriptor ¶ added in v1.10.7
func (CleanPodPolicy) Descriptor() protoreflect.EnumDescriptor
func (CleanPodPolicy) Enum ¶ added in v1.10.7
func (x CleanPodPolicy) Enum() *CleanPodPolicy
func (CleanPodPolicy) EnumDescriptor
deprecated
func (CleanPodPolicy) EnumDescriptor() ([]byte, []int)
Deprecated: Use CleanPodPolicy.Descriptor instead.
func (CleanPodPolicy) Number ¶ added in v1.10.7
func (x CleanPodPolicy) Number() protoreflect.EnumNumber
func (CleanPodPolicy) String ¶
func (x CleanPodPolicy) String() string
func (CleanPodPolicy) Type ¶ added in v1.10.7
func (CleanPodPolicy) Type() protoreflect.EnumType
type CommonReplicaSpec ¶ added in v1.13.0
type CommonReplicaSpec = plugins.CommonReplicaSpec
type DistributedMPITrainingReplicaSpec ¶
type DistributedMPITrainingReplicaSpec struct { // 1~4 deprecated. Use common instead. // Number of replicas // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/mpi.proto. Replicas int32 `protobuf:"varint,1,opt,name=replicas,proto3" json:"replicas,omitempty"` // Image used for the replica group // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/mpi.proto. Image string `protobuf:"bytes,2,opt,name=image,proto3" json:"image,omitempty"` // Resources required for the replica group // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/mpi.proto. Resources *core.Resources `protobuf:"bytes,3,opt,name=resources,proto3" json:"resources,omitempty"` // Restart policy determines whether pods will be restarted when they exit // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/mpi.proto. RestartPolicy plugins.RestartPolicy `` /* 137-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"` // The common replica spec Common *plugins.CommonReplicaSpec `protobuf:"bytes,6,opt,name=common,proto3" json:"common,omitempty"` // contains filtered or unexported fields }
Replica specification for distributed MPI training
func (*DistributedMPITrainingReplicaSpec) Descriptor
deprecated
func (*DistributedMPITrainingReplicaSpec) Descriptor() ([]byte, []int)
Deprecated: Use DistributedMPITrainingReplicaSpec.ProtoReflect.Descriptor instead.
func (*DistributedMPITrainingReplicaSpec) GetCommand ¶
func (x *DistributedMPITrainingReplicaSpec) GetCommand() []string
func (*DistributedMPITrainingReplicaSpec) GetCommon ¶ added in v1.13.0
func (x *DistributedMPITrainingReplicaSpec) GetCommon() *plugins.CommonReplicaSpec
func (*DistributedMPITrainingReplicaSpec) GetImage
deprecated
func (x *DistributedMPITrainingReplicaSpec) GetImage() string
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/mpi.proto.
func (*DistributedMPITrainingReplicaSpec) GetReplicas
deprecated
func (x *DistributedMPITrainingReplicaSpec) GetReplicas() int32
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/mpi.proto.
func (*DistributedMPITrainingReplicaSpec) GetResources
deprecated
func (x *DistributedMPITrainingReplicaSpec) GetResources() *core.Resources
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/mpi.proto.
func (*DistributedMPITrainingReplicaSpec) GetRestartPolicy
deprecated
func (x *DistributedMPITrainingReplicaSpec) GetRestartPolicy() plugins.RestartPolicy
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/mpi.proto.
func (*DistributedMPITrainingReplicaSpec) ProtoMessage ¶
func (*DistributedMPITrainingReplicaSpec) ProtoMessage()
func (*DistributedMPITrainingReplicaSpec) ProtoReflect ¶ added in v1.10.7
func (x *DistributedMPITrainingReplicaSpec) ProtoReflect() protoreflect.Message
func (*DistributedMPITrainingReplicaSpec) Reset ¶
func (x *DistributedMPITrainingReplicaSpec) Reset()
func (*DistributedMPITrainingReplicaSpec) String ¶
func (x *DistributedMPITrainingReplicaSpec) String() string
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"` // contains filtered or unexported fields }
Proto for plugin that enables distributed training using https://github.com/kubeflow/mpi-operator
func (*DistributedMPITrainingTask) Descriptor
deprecated
func (*DistributedMPITrainingTask) Descriptor() ([]byte, []int)
Deprecated: Use DistributedMPITrainingTask.ProtoReflect.Descriptor instead.
func (*DistributedMPITrainingTask) GetLauncherReplicas ¶
func (x *DistributedMPITrainingTask) GetLauncherReplicas() *DistributedMPITrainingReplicaSpec
func (*DistributedMPITrainingTask) GetRunPolicy ¶
func (x *DistributedMPITrainingTask) GetRunPolicy() *RunPolicy
func (*DistributedMPITrainingTask) GetSlots ¶
func (x *DistributedMPITrainingTask) GetSlots() int32
func (*DistributedMPITrainingTask) GetWorkerReplicas ¶
func (x *DistributedMPITrainingTask) GetWorkerReplicas() *DistributedMPITrainingReplicaSpec
func (*DistributedMPITrainingTask) ProtoMessage ¶
func (*DistributedMPITrainingTask) ProtoMessage()
func (*DistributedMPITrainingTask) ProtoReflect ¶ added in v1.10.7
func (x *DistributedMPITrainingTask) ProtoReflect() protoreflect.Message
func (*DistributedMPITrainingTask) Reset ¶
func (x *DistributedMPITrainingTask) Reset()
func (*DistributedMPITrainingTask) String ¶
func (x *DistributedMPITrainingTask) String() string
type DistributedPyTorchTrainingReplicaSpec ¶
type DistributedPyTorchTrainingReplicaSpec struct { // 1~4 deprecated. Use common instead. // Number of replicas // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/pytorch.proto. Replicas int32 `protobuf:"varint,1,opt,name=replicas,proto3" json:"replicas,omitempty"` // Image used for the replica group // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/pytorch.proto. Image string `protobuf:"bytes,2,opt,name=image,proto3" json:"image,omitempty"` // Resources required for the replica group // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/pytorch.proto. Resources *core.Resources `protobuf:"bytes,3,opt,name=resources,proto3" json:"resources,omitempty"` // Restart policy determines whether pods will be restarted when they exit // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/pytorch.proto. RestartPolicy plugins.RestartPolicy `` /* 137-byte string literal not displayed */ // The common replica spec Common *plugins.CommonReplicaSpec `protobuf:"bytes,5,opt,name=common,proto3" json:"common,omitempty"` // contains filtered or unexported fields }
func (*DistributedPyTorchTrainingReplicaSpec) Descriptor
deprecated
func (*DistributedPyTorchTrainingReplicaSpec) Descriptor() ([]byte, []int)
Deprecated: Use DistributedPyTorchTrainingReplicaSpec.ProtoReflect.Descriptor instead.
func (*DistributedPyTorchTrainingReplicaSpec) GetCommon ¶ added in v1.13.0
func (x *DistributedPyTorchTrainingReplicaSpec) GetCommon() *plugins.CommonReplicaSpec
func (*DistributedPyTorchTrainingReplicaSpec) GetImage
deprecated
func (x *DistributedPyTorchTrainingReplicaSpec) GetImage() string
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/pytorch.proto.
func (*DistributedPyTorchTrainingReplicaSpec) GetReplicas
deprecated
func (x *DistributedPyTorchTrainingReplicaSpec) GetReplicas() int32
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/pytorch.proto.
func (*DistributedPyTorchTrainingReplicaSpec) GetResources
deprecated
func (x *DistributedPyTorchTrainingReplicaSpec) GetResources() *core.Resources
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/pytorch.proto.
func (*DistributedPyTorchTrainingReplicaSpec) GetRestartPolicy
deprecated
func (x *DistributedPyTorchTrainingReplicaSpec) GetRestartPolicy() plugins.RestartPolicy
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/pytorch.proto.
func (*DistributedPyTorchTrainingReplicaSpec) ProtoMessage ¶
func (*DistributedPyTorchTrainingReplicaSpec) ProtoMessage()
func (*DistributedPyTorchTrainingReplicaSpec) ProtoReflect ¶ added in v1.10.7
func (x *DistributedPyTorchTrainingReplicaSpec) ProtoReflect() protoreflect.Message
func (*DistributedPyTorchTrainingReplicaSpec) Reset ¶
func (x *DistributedPyTorchTrainingReplicaSpec) Reset()
func (*DistributedPyTorchTrainingReplicaSpec) String ¶
func (x *DistributedPyTorchTrainingReplicaSpec) String() string
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"` // contains filtered or unexported fields }
Proto for plugin that enables distributed training using https://github.com/kubeflow/pytorch-operator
func (*DistributedPyTorchTrainingTask) Descriptor
deprecated
func (*DistributedPyTorchTrainingTask) Descriptor() ([]byte, []int)
Deprecated: Use DistributedPyTorchTrainingTask.ProtoReflect.Descriptor instead.
func (*DistributedPyTorchTrainingTask) GetElasticConfig ¶
func (x *DistributedPyTorchTrainingTask) GetElasticConfig() *ElasticConfig
func (*DistributedPyTorchTrainingTask) GetMasterReplicas ¶
func (x *DistributedPyTorchTrainingTask) GetMasterReplicas() *DistributedPyTorchTrainingReplicaSpec
func (*DistributedPyTorchTrainingTask) GetRunPolicy ¶
func (x *DistributedPyTorchTrainingTask) GetRunPolicy() *RunPolicy
func (*DistributedPyTorchTrainingTask) GetWorkerReplicas ¶
func (x *DistributedPyTorchTrainingTask) GetWorkerReplicas() *DistributedPyTorchTrainingReplicaSpec
func (*DistributedPyTorchTrainingTask) ProtoMessage ¶
func (*DistributedPyTorchTrainingTask) ProtoMessage()
func (*DistributedPyTorchTrainingTask) ProtoReflect ¶ added in v1.10.7
func (x *DistributedPyTorchTrainingTask) ProtoReflect() protoreflect.Message
func (*DistributedPyTorchTrainingTask) Reset ¶
func (x *DistributedPyTorchTrainingTask) Reset()
func (*DistributedPyTorchTrainingTask) String ¶
func (x *DistributedPyTorchTrainingTask) String() string
type DistributedTensorflowTrainingReplicaSpec ¶
type DistributedTensorflowTrainingReplicaSpec struct { // 1~4 deprecated. Use common instead. // Number of replicas // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/tensorflow.proto. Replicas int32 `protobuf:"varint,1,opt,name=replicas,proto3" json:"replicas,omitempty"` // Image used for the replica group // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/tensorflow.proto. Image string `protobuf:"bytes,2,opt,name=image,proto3" json:"image,omitempty"` // Resources required for the replica group // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/tensorflow.proto. Resources *core.Resources `protobuf:"bytes,3,opt,name=resources,proto3" json:"resources,omitempty"` // Restart policy determines whether pods will be restarted when they exit // // Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/tensorflow.proto. RestartPolicy plugins.RestartPolicy `` /* 137-byte string literal not displayed */ // The common replica spec Common *plugins.CommonReplicaSpec `protobuf:"bytes,5,opt,name=common,proto3" json:"common,omitempty"` // contains filtered or unexported fields }
func (*DistributedTensorflowTrainingReplicaSpec) Descriptor
deprecated
func (*DistributedTensorflowTrainingReplicaSpec) Descriptor() ([]byte, []int)
Deprecated: Use DistributedTensorflowTrainingReplicaSpec.ProtoReflect.Descriptor instead.
func (*DistributedTensorflowTrainingReplicaSpec) GetCommon ¶ added in v1.13.0
func (x *DistributedTensorflowTrainingReplicaSpec) GetCommon() *plugins.CommonReplicaSpec
func (*DistributedTensorflowTrainingReplicaSpec) GetImage
deprecated
func (x *DistributedTensorflowTrainingReplicaSpec) GetImage() string
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/tensorflow.proto.
func (*DistributedTensorflowTrainingReplicaSpec) GetReplicas
deprecated
func (x *DistributedTensorflowTrainingReplicaSpec) GetReplicas() int32
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/tensorflow.proto.
func (*DistributedTensorflowTrainingReplicaSpec) GetResources
deprecated
func (x *DistributedTensorflowTrainingReplicaSpec) GetResources() *core.Resources
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/tensorflow.proto.
func (*DistributedTensorflowTrainingReplicaSpec) GetRestartPolicy
deprecated
func (x *DistributedTensorflowTrainingReplicaSpec) GetRestartPolicy() plugins.RestartPolicy
Deprecated: Marked as deprecated in flyteidl/plugins/kubeflow/tensorflow.proto.
func (*DistributedTensorflowTrainingReplicaSpec) ProtoMessage ¶
func (*DistributedTensorflowTrainingReplicaSpec) ProtoMessage()
func (*DistributedTensorflowTrainingReplicaSpec) ProtoReflect ¶ added in v1.10.7
func (x *DistributedTensorflowTrainingReplicaSpec) ProtoReflect() protoreflect.Message
func (*DistributedTensorflowTrainingReplicaSpec) Reset ¶
func (x *DistributedTensorflowTrainingReplicaSpec) Reset()
func (*DistributedTensorflowTrainingReplicaSpec) String ¶
func (x *DistributedTensorflowTrainingReplicaSpec) String() string
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"` // Evaluator replicas spec EvaluatorReplicas *DistributedTensorflowTrainingReplicaSpec `protobuf:"bytes,5,opt,name=evaluator_replicas,json=evaluatorReplicas,proto3" json:"evaluator_replicas,omitempty"` // contains filtered or unexported fields }
Proto for plugin that enables distributed training using https://github.com/kubeflow/tf-operator
func (*DistributedTensorflowTrainingTask) Descriptor
deprecated
func (*DistributedTensorflowTrainingTask) Descriptor() ([]byte, []int)
Deprecated: Use DistributedTensorflowTrainingTask.ProtoReflect.Descriptor instead.
func (*DistributedTensorflowTrainingTask) GetChiefReplicas ¶
func (x *DistributedTensorflowTrainingTask) GetChiefReplicas() *DistributedTensorflowTrainingReplicaSpec
func (*DistributedTensorflowTrainingTask) GetEvaluatorReplicas ¶ added in v1.9.16
func (x *DistributedTensorflowTrainingTask) GetEvaluatorReplicas() *DistributedTensorflowTrainingReplicaSpec
func (*DistributedTensorflowTrainingTask) GetPsReplicas ¶
func (x *DistributedTensorflowTrainingTask) GetPsReplicas() *DistributedTensorflowTrainingReplicaSpec
func (*DistributedTensorflowTrainingTask) GetRunPolicy ¶
func (x *DistributedTensorflowTrainingTask) GetRunPolicy() *RunPolicy
func (*DistributedTensorflowTrainingTask) GetWorkerReplicas ¶
func (x *DistributedTensorflowTrainingTask) GetWorkerReplicas() *DistributedTensorflowTrainingReplicaSpec
func (*DistributedTensorflowTrainingTask) ProtoMessage ¶
func (*DistributedTensorflowTrainingTask) ProtoMessage()
func (*DistributedTensorflowTrainingTask) ProtoReflect ¶ added in v1.10.7
func (x *DistributedTensorflowTrainingTask) ProtoReflect() protoreflect.Message
func (*DistributedTensorflowTrainingTask) Reset ¶
func (x *DistributedTensorflowTrainingTask) Reset()
func (*DistributedTensorflowTrainingTask) String ¶
func (x *DistributedTensorflowTrainingTask) String() string
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"` // contains filtered or unexported fields }
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
deprecated
func (*ElasticConfig) Descriptor() ([]byte, []int)
Deprecated: Use ElasticConfig.ProtoReflect.Descriptor instead.
func (*ElasticConfig) GetMaxReplicas ¶
func (x *ElasticConfig) GetMaxReplicas() int32
func (*ElasticConfig) GetMaxRestarts ¶
func (x *ElasticConfig) GetMaxRestarts() int32
func (*ElasticConfig) GetMinReplicas ¶
func (x *ElasticConfig) GetMinReplicas() int32
func (*ElasticConfig) GetNprocPerNode ¶
func (x *ElasticConfig) GetNprocPerNode() int32
func (*ElasticConfig) GetRdzvBackend ¶
func (x *ElasticConfig) GetRdzvBackend() string
func (*ElasticConfig) ProtoMessage ¶
func (*ElasticConfig) ProtoMessage()
func (*ElasticConfig) ProtoReflect ¶ added in v1.10.7
func (x *ElasticConfig) ProtoReflect() protoreflect.Message
func (*ElasticConfig) Reset ¶
func (x *ElasticConfig) Reset()
func (*ElasticConfig) String ¶
func (x *ElasticConfig) String() string
type RestartPolicy ¶
type RestartPolicy = plugins.RestartPolicy
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"` // contains filtered or unexported fields }
func (*RunPolicy) Descriptor
deprecated
func (*RunPolicy) GetActiveDeadlineSeconds ¶
func (*RunPolicy) GetBackoffLimit ¶
func (*RunPolicy) GetCleanPodPolicy ¶
func (x *RunPolicy) GetCleanPodPolicy() CleanPodPolicy
func (*RunPolicy) GetTtlSecondsAfterFinished ¶
func (*RunPolicy) ProtoMessage ¶
func (*RunPolicy) ProtoMessage()
func (*RunPolicy) ProtoReflect ¶ added in v1.10.7
func (x *RunPolicy) ProtoReflect() protoreflect.Message