v1alpha2

package
v0.6.1 Latest Latest
Warning

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

 Go to latest Published: Aug 30, 2021 License: Apache-2.0 Imports: 32 Imported by: 0

Documentation

Overview

Package v1alpha2 contains API Schema definitions for the serving v1alpha2 API group +k8s:openapi-gen=true +k8s:deepcopy-gen=package,register +k8s:conversion-gen=kfserving/pkg/apis/serving +k8s:defaulter-gen=TypeMeta +groupName=serving.kubeflow.org

Package v1alpha2 contains API Schema definitions for the serving v1alpha1 API group +k8s:openapi-gen=true +k8s:deepcopy-gen=package,register +k8s:conversion-gen=kfserving/pkg/apis/serving +k8s:defaulter-gen=TypeMeta +groupName=serving.kubeflow.org

Index

Constants

View Source 
const (
	PredictorConfigKeyName   = "predictors"
	TransformerConfigKeyName = "transformers"
	ExplainerConfigKeyName   = "explainers"
)

ConfigMap Keys

View Source 
const (
	// RoutesReady is set when network configuration has completed.
	RoutesReady apis.ConditionType = "RoutesReady"
	// DefaultPredictorReady is set when default predictor has reported readiness.
	DefaultPredictorReady apis.ConditionType = "DefaultPredictorReady"
	// CanaryPredictorReady is set when canary predictor has reported readiness.
	CanaryPredictorReady apis.ConditionType = "CanaryPredictorReady"
	// DefaultTransformerReady is set when default transformer has reported readiness.
	DefaultTransformerReady apis.ConditionType = "DefaultTransformerReady"
	// CanaryTransformerReady is set when canary transformer has reported readiness.
	CanaryTransformerReady apis.ConditionType = "CanaryTransformerReady"

	// DefaultExplainerReady is set when default explainer has reported readiness.
	DefaultExplainerReady apis.ConditionType = "DefaultExplainerReady"
	// CanaryExplainerReady is set when canary explainer has reported readiness.
	CanaryExplainerReady apis.ConditionType = "CanaryExplainerReady"
)

ConditionType represents a Service condition value

View Source 
const (
	MinReplicasShouldBeLessThanMaxError = "MinReplicas cannot be greater than MaxReplicas."
	MinReplicasLowerBoundExceededError  = "MinReplicas cannot be less than 0."
	MaxReplicasLowerBoundExceededError  = "MaxReplicas cannot be less than 0."
	ParallelismLowerBoundExceededError  = "Parallelism cannot be less than 0."
	TrafficBoundsExceededError          = "TrafficPercent must be between [0, 100]."
	TrafficProvidedWithoutCanaryError   = "Canary must be specified when CanaryTrafficPercent > 0."
	UnsupportedStorageURIFormatError    = "" /* 156-byte string literal not displayed */
	InvalidISVCNameFormatError          = "" /* 235-byte string literal not displayed */
)

Known error messages

View Source 
const (
	// ExactlyOneExplainerViolatedError is a known error message
	ExactlyOneExplainerViolatedError = "Exactly one of [Custom, Alibi, AIX] must be specified in ExplainerSpec"
)
View Source 
const (
	// ExactlyOnePredictorViolatedError is a known error message
	ExactlyOnePredictorViolatedError = "Exactly one of [Custom, ONNX, Tensorflow, Triton, SKLearn, XGBoost, LightGBM] must be specified in PredictorSpec"
)
View Source 
const (
	ExactlyOneTransformerViolatedError = "Exactly one of [Custom, Feast] must be specified in TransformerSpec"
)

Constants

View Source 
const (
	InvalidLoggerType = "Invalid logger type"
)
View Source 
const (
	IsvcNameFmt string = "[a-z]([-a-z0-9]*[a-z0-9])?"
)

Validation for isvc name

Variables

View Source 
var (
	ONNXServingRestPort            = "8080"
	ONNXServingGRPCPort            = "9000"
	ONNXFileExt                    = ".onnx"
	DefaultONNXFileName            = "model.onnx"
	InvalidONNXRuntimeVersionError = "ONNX RuntimeVersion must be one of %s"
)
View Source 
var (
	DefaultPyTorchModelClassName     = "PyTorchModel"
	PyTorchServingGPUSuffix          = "-gpu"
	InvalidPyTorchRuntimeIncludesGPU = "PyTorch RuntimeVersion is not GPU enabled but GPU resources are requested. "
	InvalidPyTorchRuntimeExcludesGPU = "PyTorch RuntimeVersion is GPU enabled but GPU resources are not requested. "
)
View Source 
var (
	TensorflowEntrypointCommand         = "/usr/bin/tensorflow_model_server"
	TensorflowServingGRPCPort           = "9000"
	TensorflowServingRestPort           = "8080"
	TensorflowServingGPUSuffix          = "-gpu"
	InvalidTensorflowRuntimeIncludesGPU = "Tensorflow RuntimeVersion is not GPU enabled but GPU resources are requested. "
	InvalidTensorflowRuntimeExcludesGPU = "Tensorflow RuntimeVersion is GPU enabled but GPU resources are not requested. "
)
View Source 
var (
	// For versioning see https://github.com/triton-inference-server/server/releases
	TritonISGRPCPort = int32(9000)
	TritonISRestPort = int32(8080)
)
View Source 
var (
	SupportedStorageURIPrefixList = []string{"gs://", "s3://", "pvc://", "file://", "https://", "http://"}
	AzureBlobURL                  = "blob.core.windows.net"
	AzureBlobURIRegEx             = "https://(.+?).blob.core.windows.net/(.+)"
	IsvcRegexp                    = regexp.MustCompile("^" + IsvcNameFmt + "$")
)
View Source 
var (
	// APIVersion is the current API version used to register these objects
	APIVersion = "v1alpha2"

	// SchemeGroupVersion is group version used to register these objects
	SchemeGroupVersion = schema.GroupVersion{Group: constants.KFServingAPIGroupName, Version: APIVersion}

	// SchemeBuilder is used to add go types to the GroupVersionKind scheme
	SchemeBuilder = &scheme.Builder{GroupVersion: SchemeGroupVersion}

	// AddToScheme is required by pkg/client/...
	AddToScheme = SchemeBuilder.AddToScheme
)
View Source 
var (
	InvalidSKLearnRuntimeVersionError = "SKLearn RuntimeVersion must be one of %s"
)

Functions

func GetIntReference added in v0.3.0 

func GetIntReference(number int) *int

GetIntReference returns the pointer for the integer input

func GetOpenAPIDefinitions 

func GetOpenAPIDefinitions(ref common.ReferenceCallback) map[string]common.OpenAPIDefinition

func Resource 

func Resource(resource string) schema.GroupResource

Resource is required by pkg/client/listers/...

Types

type AIXExplainerSpec added in v0.5.0 

type AIXExplainerSpec struct {
	// The type of AIX explainer
	Type AIXExplainerType `json:"type"`
	// The location of a trained explanation model
	StorageURI string `json:"storageUri,omitempty"`
	// Defaults to latest AIX Version
	RuntimeVersion string `json:"runtimeVersion,omitempty"`
	// Defaults to requests and limits of 1CPU, 2Gb MEM.
	Resources v1.ResourceRequirements `json:"resources,omitempty"`
	// Inline custom parameter settings for explainer
	Config map[string]string `json:"config,omitempty"`
}

AIXExplainerSpec defines the arguments for configuring an AIX Explanation Server

func (*AIXExplainerSpec) ApplyDefaults added in v0.5.0 

func (s *AIXExplainerSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*AIXExplainerSpec) CreateExplainerContainer added in v0.5.0 

func (s *AIXExplainerSpec) CreateExplainerContainer(modelName string, parallelism int, predictorHost string, config *InferenceServicesConfig) *v1.Container

func (*AIXExplainerSpec) DeepCopy added in v0.5.0 

func (in *AIXExplainerSpec) DeepCopy() *AIXExplainerSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new AIXExplainerSpec.

func (*AIXExplainerSpec) DeepCopyInto added in v0.5.0 

func (in *AIXExplainerSpec) DeepCopyInto(out *AIXExplainerSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*AIXExplainerSpec) GetResourceRequirements added in v0.5.0 

func (s *AIXExplainerSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*AIXExplainerSpec) GetStorageUri added in v0.5.0 

func (s *AIXExplainerSpec) GetStorageUri() string

func (*AIXExplainerSpec) Validate added in v0.5.0 

func (s *AIXExplainerSpec) Validate(config *InferenceServicesConfig) error

type AIXExplainerType added in v0.5.0 

type AIXExplainerType string
const (
	AIXLimeImageExplainer AIXExplainerType = "LimeImages"
)

type AlibiExplainerSpec 

type AlibiExplainerSpec struct {
	// The type of Alibi explainer
	Type AlibiExplainerType `json:"type"`
	// The location of a trained explanation model
	StorageURI string `json:"storageUri,omitempty"`
	// Alibi docker image version which defaults to latest release
	RuntimeVersion string `json:"runtimeVersion,omitempty"`
	// Defaults to requests and limits of 1CPU, 2Gb MEM.
	Resources v1.ResourceRequirements `json:"resources,omitempty"`
	// Inline custom parameter settings for explainer
	Config map[string]string `json:"config,omitempty"`
}

AlibiExplainerSpec defines the arguments for configuring an Alibi Explanation Server

func (*AlibiExplainerSpec) ApplyDefaults 

func (s *AlibiExplainerSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*AlibiExplainerSpec) CreateExplainerContainer added in v0.3.0 

func (s *AlibiExplainerSpec) CreateExplainerContainer(modelName string, parallelism int, predictorHost string, config *InferenceServicesConfig) *v1.Container

func (*AlibiExplainerSpec) DeepCopy 

func (in *AlibiExplainerSpec) DeepCopy() *AlibiExplainerSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new AlibiExplainerSpec.

func (*AlibiExplainerSpec) DeepCopyInto 

func (in *AlibiExplainerSpec) DeepCopyInto(out *AlibiExplainerSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*AlibiExplainerSpec) GetResourceRequirements added in v0.3.0 

func (s *AlibiExplainerSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*AlibiExplainerSpec) GetStorageUri 

func (s *AlibiExplainerSpec) GetStorageUri() string

func (*AlibiExplainerSpec) Validate 

func (s *AlibiExplainerSpec) Validate(config *InferenceServicesConfig) error

type AlibiExplainerType 

type AlibiExplainerType string
const (
	AlibiAnchorsTabularExplainer  AlibiExplainerType = "AnchorTabular"
	AlibiAnchorsImageExplainer    AlibiExplainerType = "AnchorImages"
	AlibiAnchorsTextExplainer     AlibiExplainerType = "AnchorText"
	AlibiCounterfactualsExplainer AlibiExplainerType = "Counterfactuals"
	AlibiContrastiveExplainer     AlibiExplainerType = "Contrastive"
)

type Batcher added in v0.4.0 

type Batcher struct {
	// MaxBatchSize of batcher service
	// +optional
	MaxBatchSize *int `json:"maxBatchSize,omitempty"`
	// MaxLatency of batcher service
	// +optional
	MaxLatency *int `json:"maxLatency,omitempty"`
	// Timeout of batcher service
	// +optional
	Timeout *int `json:"timeout,omitempty"`
}

Batcher provides optional payload batcher for all endpoints +experimental

func (*Batcher) DeepCopy added in v0.4.0 

func (in *Batcher) DeepCopy() *Batcher

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new Batcher.

func (*Batcher) DeepCopyInto added in v0.4.0 

func (in *Batcher) DeepCopyInto(out *Batcher)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

type ComponentStatusMap added in v0.3.0 

type ComponentStatusMap *map[constants.InferenceServiceComponent]StatusConfigurationSpec

EndpointStatusMap defines the observed state of InferenceService endpoints

type CustomSpec 

type CustomSpec struct {
	Container v1.Container `json:"container"`
}

CustomSpec provides a hook for arbitrary container configuration.

func (*CustomSpec) ApplyDefaults 

func (c *CustomSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*CustomSpec) CreateExplainerContainer added in v0.3.0 

func (c *CustomSpec) CreateExplainerContainer(modelName string, parallelism int, predictUrl string, config *InferenceServicesConfig) *v1.Container

func (*CustomSpec) DeepCopy 

func (in *CustomSpec) DeepCopy() *CustomSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new CustomSpec.

func (*CustomSpec) DeepCopyInto 

func (in *CustomSpec) DeepCopyInto(out *CustomSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*CustomSpec) GetContainer added in v0.3.0 

func (c *CustomSpec) GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container

func (*CustomSpec) GetContainerSpec added in v0.3.0 

func (c *CustomSpec) GetContainerSpec() *v1.Container

GetContainerSpec for the CustomSpec

func (*CustomSpec) GetResourceRequirements added in v0.3.0 

func (c *CustomSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*CustomSpec) GetStorageUri 

func (c *CustomSpec) GetStorageUri() string

func (*CustomSpec) Validate 

func (c *CustomSpec) Validate(config *InferenceServicesConfig) error

type DeploymentSpec 

type DeploymentSpec struct {
	// ServiceAccountName is the name of the ServiceAccount to use to run the service
	// +optional
	ServiceAccountName string `json:"serviceAccountName,omitempty"`
	// Minimum number of replicas which defaults to 1, when minReplicas = 0 pods scale down to 0 in case of no traffic
	// +optional
	MinReplicas *int `json:"minReplicas,omitempty"`
	// This is the up bound for autoscaler to scale to
	// +optional
	MaxReplicas int `json:"maxReplicas,omitempty"`
	// Parallelism specifies how many requests can be processed concurrently, this sets the hard limit of the container
	// concurrency(https://knative.dev/docs/serving/autoscaling/concurrency).
	Parallelism int `json:"parallelism,omitempty"`
	// Activate request/response logging
	// +optional
	Logger *Logger `json:"logger,omitempty"`
	// Activate request batching
	// +optional
	Batcher *Batcher `json:"batcher,omitempty"`
}

DeploymentSpec defines the configuration for a given InferenceService service component

func (*DeploymentSpec) DeepCopy 

func (in *DeploymentSpec) DeepCopy() *DeploymentSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new DeploymentSpec.

func (*DeploymentSpec) DeepCopyInto 

func (in *DeploymentSpec) DeepCopyInto(out *DeploymentSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

type EndpointSpec 

type EndpointSpec struct {
	// Predictor defines the model serving spec
	// +required
	Predictor PredictorSpec `json:"predictor"`

	// Explainer defines the model explanation service spec,
	// explainer service calls to predictor or transformer if it is specified.
	// +optional
	Explainer *ExplainerSpec `json:"explainer,omitempty"`

	// Transformer defines the pre/post processing before and after the predictor call,
	// transformer service calls to predictor service.
	// +optional
	Transformer *TransformerSpec `json:"transformer,omitempty"`
}

func (*EndpointSpec) DeepCopy 

func (in *EndpointSpec) DeepCopy() *EndpointSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new EndpointSpec.

func (*EndpointSpec) DeepCopyInto 

func (in *EndpointSpec) DeepCopyInto(out *EndpointSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

type Explainer added in v0.3.0 

type Explainer interface {
	GetResourceRequirements() *v1.ResourceRequirements
	GetStorageUri() string
	CreateExplainerContainer(modelName string, parallelism int, predictorHost string, config *InferenceServicesConfig) *v1.Container
	ApplyDefaults(config *InferenceServicesConfig)
	Validate(config *InferenceServicesConfig) error
}

+kubebuilder:object:generate=false

type ExplainerConfig 

type ExplainerConfig struct {
	ContainerImage string `json:"image"`

	DefaultImageVersion string `json:"defaultImageVersion"`
}

+kubebuilder:object:generate=false

type ExplainerSpec 

type ExplainerSpec struct {
	// Spec for alibi explainer
	Alibi *AlibiExplainerSpec `json:"alibi,omitempty"`
	// Spec for AIX explainer
	AIX *AIXExplainerSpec `json:"aix,omitempty"`
	// Spec for a custom explainer
	Custom *CustomSpec `json:"custom,omitempty"`

	DeploymentSpec `json:",inline"`
}

ExplainerSpec defines the arguments for a model explanation server, The following fields follow a "1-of" semantic. Users must specify exactly one spec.

func (*ExplainerSpec) ApplyDefaults 

func (e *ExplainerSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*ExplainerSpec) CreateExplainerContainer added in v0.3.0 

func (e *ExplainerSpec) CreateExplainerContainer(modelName string, parallelism int, predictorHost string, config *InferenceServicesConfig) *v1.Container

func (*ExplainerSpec) DeepCopy 

func (in *ExplainerSpec) DeepCopy() *ExplainerSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new ExplainerSpec.

func (*ExplainerSpec) DeepCopyInto 

func (in *ExplainerSpec) DeepCopyInto(out *ExplainerSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*ExplainerSpec) GetStorageUri 

func (e *ExplainerSpec) GetStorageUri() string

Returns a URI to the explainer. This URI is passed to the model-initializer via the ModelInitializerSourceUriInternalAnnotationKey

func (*ExplainerSpec) Validate 

func (e *ExplainerSpec) Validate(config *InferenceServicesConfig) error

type ExplainersConfig 

type ExplainersConfig struct {
	AlibiExplainer ExplainerConfig `json:"alibi,omitempty"`
	AIXExplainer   ExplainerConfig `json:"aix,omitempty"`
}

+kubebuilder:object:generate=false

type InferenceService added in v0.3.0 

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

	Spec   InferenceServiceSpec   `json:"spec,omitempty"`
	Status InferenceServiceStatus `json:"status,omitempty"`
}

InferenceService is the Schema for the services API +k8s:openapi-gen=true +kubebuilder:object:root=true +kubebuilder:subresource:status +kubebuilder:printcolumn:name="URL",type="string",JSONPath=".status.url" +kubebuilder:printcolumn:name="Ready",type="string",JSONPath=".status.conditions[?(@.type=='Ready')].status" +kubebuilder:printcolumn:name="Default Traffic",type="integer",JSONPath=".status.traffic" +kubebuilder:printcolumn:name="Canary Traffic",type="integer",JSONPath=".status.canaryTraffic" +kubebuilder:printcolumn:name="Age",type="date",JSONPath=".metadata.creationTimestamp" +kubebuilder:resource:path=inferenceservices,shortName=isvc

func (*InferenceService) ConvertFrom added in v0.5.0 

func (dst *InferenceService) ConvertFrom(srcRaw conversion.Hub) error

Convert from hub version v1beta1 to v1alpha2

func (*InferenceService) ConvertTo added in v0.5.0 

func (src *InferenceService) ConvertTo(dstRaw conversion.Hub) error

Convert to hub version from v1alpha2 to v1beta1

func (*InferenceService) DeepCopy added in v0.3.0 

func (in *InferenceService) DeepCopy() *InferenceService

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new InferenceService.

func (*InferenceService) DeepCopyInto added in v0.3.0 

func (in *InferenceService) DeepCopyInto(out *InferenceService)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*InferenceService) DeepCopyObject added in v0.3.0 

func (in *InferenceService) DeepCopyObject() runtime.Object

DeepCopyObject is an autogenerated deepcopy function, copying the receiver, creating a new runtime.Object.

func (*InferenceService) Default added in v0.3.0 

func (isvc *InferenceService) Default()

func (*InferenceService) ValidateCreate added in v0.3.0 

func (isvc *InferenceService) ValidateCreate() error

ValidateCreate implements https://godoc.org/sigs.k8s.io/controller-runtime/pkg/webhook/admission#Validator

func (*InferenceService) ValidateDelete added in v0.5.0 

func (isvc *InferenceService) ValidateDelete() error

ValidateDelete implements webhook.Validator so a webhook will be registered for the type

func (*InferenceService) ValidateUpdate added in v0.3.0 

func (isvc *InferenceService) ValidateUpdate(old runtime.Object) error

ValidateUpdate implements https://godoc.org/sigs.k8s.io/controller-runtime/pkg/webhook/admission#Validator

type InferenceServiceList added in v0.3.0 

type InferenceServiceList struct {
	metav1.TypeMeta `json:",inline"`
	metav1.ListMeta `json:"metadata,omitempty"`
	// +listType=set
	Items []InferenceService `json:"items"`
}

+kubebuilder:object:root=true InferenceServiceList contains a list of Service

func (*InferenceServiceList) DeepCopy added in v0.3.0 

func (in *InferenceServiceList) DeepCopy() *InferenceServiceList

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new InferenceServiceList.

func (*InferenceServiceList) DeepCopyInto added in v0.3.0 

func (in *InferenceServiceList) DeepCopyInto(out *InferenceServiceList)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*InferenceServiceList) DeepCopyObject added in v0.3.0 

func (in *InferenceServiceList) DeepCopyObject() runtime.Object

DeepCopyObject is an autogenerated deepcopy function, copying the receiver, creating a new runtime.Object.

type InferenceServiceSpec added in v0.3.0 

type InferenceServiceSpec struct {
	// Default defines default InferenceService endpoints
	// +required
	Default EndpointSpec `json:"default"`
	// Canary defines alternate endpoints to route a percentage of traffic.
	// +optional
	Canary *EndpointSpec `json:"canary,omitempty"`
	// CanaryTrafficPercent defines the percentage of traffic going to canary InferenceService endpoints
	// +optional
	CanaryTrafficPercent *int `json:"canaryTrafficPercent,omitempty"`
}

InferenceServiceSpec defines the desired state of InferenceService

func (*InferenceServiceSpec) DeepCopy added in v0.3.0 

func (in *InferenceServiceSpec) DeepCopy() *InferenceServiceSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new InferenceServiceSpec.

func (*InferenceServiceSpec) DeepCopyInto added in v0.3.0 

func (in *InferenceServiceSpec) DeepCopyInto(out *InferenceServiceSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

type InferenceServiceState added in v0.4.0 

type InferenceServiceState string

InferenceState describes the Readiness of the InferenceService 

const (
	InferenceServiceReadyState    InferenceServiceState = "InferenceServiceReady"
	InferenceServiceNotReadyState InferenceServiceState = "InferenceServiceNotReady"
)

Different InferenceServiceState an InferenceService may have.

type InferenceServiceStatus added in v0.3.0 

type InferenceServiceStatus struct {
	duckv1beta1.Status `json:",inline"`
	// URL of the InferenceService
	URL string `json:"url,omitempty"`
	// Traffic percentage that goes to default services
	Traffic int `json:"traffic,omitempty"`
	// Traffic percentage that goes to canary services
	CanaryTraffic int `json:"canaryTraffic,omitempty"`
	// Statuses for the default endpoints of the InferenceService
	Default ComponentStatusMap `json:"default,omitempty"`
	// Statuses for the canary endpoints of the InferenceService
	Canary ComponentStatusMap `json:"canary,omitempty"`
	// Addressable URL for eventing
	Address *duckv1beta1.Addressable `json:"address,omitempty"`
}

InferenceServiceStatus defines the observed state of InferenceService

func (*InferenceServiceStatus) DeepCopy added in v0.3.0 

func (in *InferenceServiceStatus) DeepCopy() *InferenceServiceStatus

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new InferenceServiceStatus.

func (*InferenceServiceStatus) DeepCopyInto added in v0.3.0 

func (in *InferenceServiceStatus) DeepCopyInto(out *InferenceServiceStatus)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*InferenceServiceStatus) GetCondition added in v0.3.0 

func (ss *InferenceServiceStatus) GetCondition(t apis.ConditionType) *apis.Condition

GetCondition returns the condition by name.

func (*InferenceServiceStatus) InitializeConditions added in v0.3.0 

func (ss *InferenceServiceStatus) InitializeConditions()

func (*InferenceServiceStatus) IsReady added in v0.3.0 

func (ss *InferenceServiceStatus) IsReady() bool

IsReady returns if the service is ready to serve the requested configuration.

func (*InferenceServiceStatus) PropagateCanaryStatus added in v0.3.0 

func (ss *InferenceServiceStatus) PropagateCanaryStatus(component constants.InferenceServiceComponent,
	canaryStatus *knservingv1.ServiceStatus)

PropagateCanaryStatus propagates the status for the canary spec

func (*InferenceServiceStatus) PropagateDefaultStatus added in v0.3.0 

func (ss *InferenceServiceStatus) PropagateDefaultStatus(component constants.InferenceServiceComponent,
	defaultStatus *knservingv1.ServiceStatus)

PropagateDefaultStatus propagates the status for the default spec

func (*InferenceServiceStatus) PropagateRouteStatus added in v0.3.0 

func (ss *InferenceServiceStatus) PropagateRouteStatus(vs *VirtualServiceStatus)

PropagateRouteStatus propagates route's status to the service's status.

type InferenceServicesConfig added in v0.3.0 

type InferenceServicesConfig struct {
	Transformers *TransformersConfig `json:"transformers"`
	Predictors   *PredictorsConfig   `json:"predictors"`
	Explainers   *ExplainersConfig   `json:"explainers"`
}

+kubebuilder:object:generate=false

func GetInferenceServicesConfig added in v0.3.0 

func GetInferenceServicesConfig(client client.Client) (*InferenceServicesConfig, error)

func NewInferenceServicesConfig added in v0.3.0 

func NewInferenceServicesConfig(configMap *v1.ConfigMap) (*InferenceServicesConfig, error)

type LightGBMSpec added in v0.5.0 

type LightGBMSpec struct {
	// The URI of the trained model which contains model.bst
	StorageURI string `json:"storageUri"`
	// Number of thread to be used by LightGBM
	NThread int `json:"nthread,omitempty"`
	// LightGBM KFServer docker image version which defaults to latest release
	RuntimeVersion string `json:"runtimeVersion,omitempty"`
	// Defaults to requests and limits of 1CPU, 2Gb MEM.
	Resources v1.ResourceRequirements `json:"resources,omitempty"`
}

LightGBMSpec defines arguments for configuring LightGBM model serving.

func (*LightGBMSpec) ApplyDefaults added in v0.5.0 

func (x *LightGBMSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*LightGBMSpec) DeepCopy added in v0.5.0 

func (in *LightGBMSpec) DeepCopy() *LightGBMSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new LightGBMSpec.

func (*LightGBMSpec) DeepCopyInto added in v0.5.0 

func (in *LightGBMSpec) DeepCopyInto(out *LightGBMSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*LightGBMSpec) GetContainer added in v0.5.0 

func (x *LightGBMSpec) GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container

func (*LightGBMSpec) GetResourceRequirements added in v0.5.0 

func (x *LightGBMSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*LightGBMSpec) GetStorageUri added in v0.5.0 

func (x *LightGBMSpec) GetStorageUri() string

func (*LightGBMSpec) Validate added in v0.5.0 

func (x *LightGBMSpec) Validate(config *InferenceServicesConfig) error

type Logger added in v0.3.0 

type Logger struct {
	// URL to send request logging CloudEvents
	// +optional
	Url *string `json:"url,omitempty"`
	// What payloads to log: [all, request, response]
	Mode LoggerMode `json:"mode,omitempty"`
}

Logger provides optional payload logging for all endpoints +experimental

func (*Logger) DeepCopy added in v0.3.0 

func (in *Logger) DeepCopy() *Logger

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new Logger.

func (*Logger) DeepCopyInto added in v0.3.0 

func (in *Logger) DeepCopyInto(out *Logger)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

type LoggerMode added in v0.3.0 

type LoggerMode string
const (
	LogAll      LoggerMode = "all"
	LogRequest  LoggerMode = "request"
	LogResponse LoggerMode = "response"
)

type ONNXSpec 

type ONNXSpec struct {
	// The URI of the exported onnx model(model.onnx)
	StorageURI string `json:"storageUri"`
	// ONNXRuntime docker image versions, default version can be set in the inferenceservice configmap
	RuntimeVersion string `json:"runtimeVersion,omitempty"`
	// Defaults to requests and limits of 1CPU, 2Gb MEM.
	Resources v1.ResourceRequirements `json:"resources,omitempty"`
}

ONNXSpec defines arguments for configuring ONNX model serving.

func (*ONNXSpec) ApplyDefaults 

func (s *ONNXSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*ONNXSpec) DeepCopy 

func (in *ONNXSpec) DeepCopy() *ONNXSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new ONNXSpec.

func (*ONNXSpec) DeepCopyInto 

func (in *ONNXSpec) DeepCopyInto(out *ONNXSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*ONNXSpec) GetContainer added in v0.3.0 

func (s *ONNXSpec) GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container

func (*ONNXSpec) GetResourceRequirements added in v0.3.0 

func (s *ONNXSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*ONNXSpec) GetStorageUri 

func (s *ONNXSpec) GetStorageUri() string

func (*ONNXSpec) Validate 

func (s *ONNXSpec) Validate(config *InferenceServicesConfig) error

type PMMLSpec added in v0.5.0 

type PMMLSpec struct {
	// The URI of the trained model which contains model.pmml
	StorageURI string `json:"storageUri"`
	// PMML KFServer docker image version which defaults to latest release
	RuntimeVersion string `json:"runtimeVersion,omitempty"`
	// Defaults to requests and limits of 1CPU, 2Gb MEM.
	Resources v1.ResourceRequirements `json:"resources,omitempty"`
}

PMMLSpec defines arguments for configuring PMML model serving.

func (*PMMLSpec) ApplyDefaults added in v0.5.0 

func (s *PMMLSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*PMMLSpec) DeepCopy added in v0.5.0 

func (in *PMMLSpec) DeepCopy() *PMMLSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new PMMLSpec.

func (*PMMLSpec) DeepCopyInto added in v0.5.0 

func (in *PMMLSpec) DeepCopyInto(out *PMMLSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*PMMLSpec) GetContainer added in v0.5.0 

func (s *PMMLSpec) GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container

func (*PMMLSpec) GetResourceRequirements added in v0.5.0 

func (s *PMMLSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*PMMLSpec) GetStorageUri added in v0.5.0 

func (s *PMMLSpec) GetStorageUri() string

func (*PMMLSpec) Validate added in v0.5.0 

func (s *PMMLSpec) Validate(config *InferenceServicesConfig) error

type Predictor added in v0.3.0 

type Predictor interface {
	GetResourceRequirements() *v1.ResourceRequirements
	GetStorageUri() string
	GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container
	ApplyDefaults(config *InferenceServicesConfig)
	Validate(config *InferenceServicesConfig) error
}

+kubebuilder:object:generate=false

type PredictorConfig added in v0.3.0 

type PredictorConfig struct {
	ContainerImage string `json:"image"`

	DefaultImageVersion    string `json:"defaultImageVersion"`
	DefaultGpuImageVersion string `json:"defaultGpuImageVersion"`
	// Default timeout of predictor for serving a request, in seconds
	DefaultTimeout int64 `json:"defaultTimeout,string,omitempty"`
}

+kubebuilder:object:generate=false

type PredictorProtocols added in v0.5.0 

type PredictorProtocols struct {
	V1 *PredictorConfig `json:"v1,omitempty"`
	V2 *PredictorConfig `json:"v2,omitempty"`
}

+kubebuilder:object:generate=false

type PredictorSpec 

type PredictorSpec struct {
	// Spec for a custom predictor
	Custom *CustomSpec `json:"custom,omitempty"`
	// Spec for Tensorflow Serving (https://github.com/tensorflow/serving)
	Tensorflow *TensorflowSpec `json:"tensorflow,omitempty"`
	// Spec for Triton Inference Server (https://github.com/triton-inference-server/server)
	Triton *TritonSpec `json:"triton,omitempty"`
	// Spec for XGBoost predictor
	XGBoost *XGBoostSpec `json:"xgboost,omitempty"`
	// Spec for LightGBM predictor
	LightGBM *LightGBMSpec `json:"lightgbm,omitempty"`
	// Spec for SKLearn predictor
	SKLearn *SKLearnSpec `json:"sklearn,omitempty"`
	// Spec for ONNX runtime (https://github.com/microsoft/onnxruntime)
	ONNX *ONNXSpec `json:"onnx,omitempty"`
	// Spec for PyTorch predictor
	PyTorch *PyTorchSpec `json:"pytorch,omitempty"`
	// Spec for PMML predictor
	PMML *PMMLSpec `json:"pmml,omitempty"`

	DeploymentSpec `json:",inline"`
}

PredictorSpec defines the configuration for a predictor, The following fields follow a "1-of" semantic. Users must specify exactly one spec.

func (*PredictorSpec) ApplyDefaults 

func (p *PredictorSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*PredictorSpec) DeepCopy 

func (in *PredictorSpec) DeepCopy() *PredictorSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new PredictorSpec.

func (*PredictorSpec) DeepCopyInto 

func (in *PredictorSpec) DeepCopyInto(out *PredictorSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*PredictorSpec) GetContainer added in v0.3.0 

func (p *PredictorSpec) GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container

func (*PredictorSpec) GetStorageUri 

func (p *PredictorSpec) GetStorageUri() string

Returns a URI to the model. This URI is passed to the storage-initializer via the StorageInitializerSourceUriInternalAnnotationKey

func (*PredictorSpec) Validate 

func (p *PredictorSpec) Validate(config *InferenceServicesConfig) error

type PredictorsConfig added in v0.3.0 

type PredictorsConfig struct {
	Tensorflow PredictorConfig    `json:"tensorflow,omitempty"`
	Triton     PredictorConfig    `json:"triton,omitempty"`
	Xgboost    PredictorProtocols `json:"xgboost,omitempty"`
	LightGBM   PredictorConfig    `json:"lightgbm,omitempty"`
	SKlearn    PredictorProtocols `json:"sklearn,omitempty"`
	PyTorch    PredictorConfig    `json:"pytorch,omitempty"`
	ONNX       PredictorConfig    `json:"onnx,omitempty"`
	PMML       PredictorConfig    `json:"pmml,omitempty"`
}

+kubebuilder:object:generate=false

type PyTorchSpec 

type PyTorchSpec struct {
	// The URI of the trained model which contains model.pt
	StorageURI string `json:"storageUri"`
	// Defaults PyTorch model class name to 'PyTorchModel'
	ModelClassName string `json:"modelClassName,omitempty"`
	// PyTorch KFServer docker image version which defaults to latest release
	RuntimeVersion string `json:"runtimeVersion,omitempty"`
	// Defaults to requests and limits of 1CPU, 2Gb MEM.
	Resources v1.ResourceRequirements `json:"resources,omitempty"`
}

PyTorchSpec defines arguments for configuring PyTorch model serving.

func (*PyTorchSpec) ApplyDefaults 

func (p *PyTorchSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*PyTorchSpec) DeepCopy 

func (in *PyTorchSpec) DeepCopy() *PyTorchSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new PyTorchSpec.

func (*PyTorchSpec) DeepCopyInto 

func (in *PyTorchSpec) DeepCopyInto(out *PyTorchSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*PyTorchSpec) GetContainer added in v0.3.0 

func (p *PyTorchSpec) GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container

func (*PyTorchSpec) GetResourceRequirements added in v0.3.0 

func (p *PyTorchSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*PyTorchSpec) GetStorageUri 

func (p *PyTorchSpec) GetStorageUri() string

func (*PyTorchSpec) Validate 

func (p *PyTorchSpec) Validate(config *InferenceServicesConfig) error

type SKLearnSpec 

type SKLearnSpec struct {
	// The URI of the trained model which contains model.pickle, model.pkl or model.joblib
	StorageURI string `json:"storageUri"`
	// SKLearn KFServer docker image version which defaults to latest release
	RuntimeVersion string `json:"runtimeVersion,omitempty"`
	// Defaults to requests and limits of 1CPU, 2Gb MEM.
	Resources v1.ResourceRequirements `json:"resources,omitempty"`
}

SKLearnSpec defines arguments for configuring SKLearn model serving.

func (*SKLearnSpec) ApplyDefaults 

func (s *SKLearnSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*SKLearnSpec) DeepCopy 

func (in *SKLearnSpec) DeepCopy() *SKLearnSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new SKLearnSpec.

func (*SKLearnSpec) DeepCopyInto 

func (in *SKLearnSpec) DeepCopyInto(out *SKLearnSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*SKLearnSpec) GetContainer added in v0.3.0 

func (s *SKLearnSpec) GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container

func (*SKLearnSpec) GetResourceRequirements added in v0.3.0 

func (s *SKLearnSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*SKLearnSpec) GetStorageUri 

func (s *SKLearnSpec) GetStorageUri() string

func (*SKLearnSpec) Validate 

func (s *SKLearnSpec) Validate(config *InferenceServicesConfig) error

type StatusConfigurationSpec 

type StatusConfigurationSpec struct {
	// Latest revision name that is in ready state
	Name string `json:"name,omitempty"`
	// Host name of the service
	Hostname string `json:"host,omitempty"`
}

StatusConfigurationSpec describes the state of the configuration receiving traffic.

func (*StatusConfigurationSpec) DeepCopy 

func (in *StatusConfigurationSpec) DeepCopy() *StatusConfigurationSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new StatusConfigurationSpec.

func (*StatusConfigurationSpec) DeepCopyInto 

func (in *StatusConfigurationSpec) DeepCopyInto(out *StatusConfigurationSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

type TensorflowSpec 

type TensorflowSpec struct {
	// The URI for the saved model(https://www.tensorflow.org/tutorials/keras/save_and_load)
	StorageURI string `json:"storageUri"`
	// TFServing docker image version(https://hub.docker.com/r/tensorflow/serving), default version can be set in the inferenceservice configmap.
	RuntimeVersion string `json:"runtimeVersion,omitempty"`
	// Defaults to requests and limits of 1CPU, 2Gb MEM.
	Resources v1.ResourceRequirements `json:"resources,omitempty"`
}

TensorflowSpec defines arguments for configuring Tensorflow model serving.

func (*TensorflowSpec) ApplyDefaults 

func (t *TensorflowSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*TensorflowSpec) DeepCopy 

func (in *TensorflowSpec) DeepCopy() *TensorflowSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new TensorflowSpec.

func (*TensorflowSpec) DeepCopyInto 

func (in *TensorflowSpec) DeepCopyInto(out *TensorflowSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*TensorflowSpec) GetContainer added in v0.3.0 

func (t *TensorflowSpec) GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container

func (*TensorflowSpec) GetResourceRequirements added in v0.3.0 

func (t *TensorflowSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*TensorflowSpec) GetStorageUri 

func (t *TensorflowSpec) GetStorageUri() string

func (*TensorflowSpec) Validate 

func (t *TensorflowSpec) Validate(config *InferenceServicesConfig) error

type Transformer added in v0.3.0 

type Transformer interface {
	GetContainerSpec() *v1.Container
	GetStorageUri() string
	ApplyDefaults(config *InferenceServicesConfig)
	Validate(config *InferenceServicesConfig) error
}

Transformer interface is implemented by all Transformers +kubebuilder:object:generate=false

type TransformerConfig added in v0.3.0 

type TransformerConfig struct {
	ContainerImage string `json:"image"`

	DefaultImageVersion string `json:"defaultImageVersion"`
}

+kubebuilder:object:generate=false

type TransformerSpec 

type TransformerSpec struct {
	// Spec for a custom transformer
	Custom *CustomSpec `json:"custom,omitempty"`

	DeploymentSpec `json:",inline"`
}

TransformerSpec defines transformer service for pre/post processing

func (*TransformerSpec) ApplyDefaults added in v0.3.0 

func (t *TransformerSpec) ApplyDefaults(config *InferenceServicesConfig)

ApplyDefaults to the TransformerSpec

func (*TransformerSpec) DeepCopy 

func (in *TransformerSpec) DeepCopy() *TransformerSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new TransformerSpec.

func (*TransformerSpec) DeepCopyInto 

func (in *TransformerSpec) DeepCopyInto(out *TransformerSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*TransformerSpec) GetContainerSpec added in v0.3.0 

func (t *TransformerSpec) GetContainerSpec(metadata metav1.ObjectMeta, isCanary bool) *v1.Container

GetContainerSpec for the transformer

func (*TransformerSpec) GetStorageUri added in v0.3.0 

func (t *TransformerSpec) GetStorageUri() string

Returns a URI to the model. This URI is passed to the storage-initializer via the StorageInitializerSourceUriInternalAnnotationKey

func (*TransformerSpec) Validate added in v0.3.0 

func (t *TransformerSpec) Validate(config *InferenceServicesConfig) error

Validate the TransformerSpec

type TransformersConfig added in v0.3.0 

type TransformersConfig struct {
	Feast TransformerConfig `json:"feast,omitempty"`
}

+kubebuilder:object:generate=false

type TritonSpec added in v0.4.0 

type TritonSpec struct {
	// The URI for the trained model repository(https://docs.nvidia.com/deeplearning/triton-inference-server/master-user-guide/docs/model_repository.html)
	StorageURI string `json:"storageUri"`
	// Triton Inference Server docker image version, default version can be set in the inferenceservice configmap
	RuntimeVersion string `json:"runtimeVersion,omitempty"`
	// Defaults to requests and limits of 1CPU, 2Gb MEM.
	Resources v1.ResourceRequirements `json:"resources,omitempty"`
}

TritonSpec defines arguments for configuring Triton Inference Server.

func (*TritonSpec) ApplyDefaults added in v0.4.0 

func (t *TritonSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*TritonSpec) DeepCopy added in v0.4.0 

func (in *TritonSpec) DeepCopy() *TritonSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new TritonSpec.

func (*TritonSpec) DeepCopyInto added in v0.4.0 

func (in *TritonSpec) DeepCopyInto(out *TritonSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*TritonSpec) GetContainer added in v0.4.0 

func (t *TritonSpec) GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container

func (*TritonSpec) GetResourceRequirements added in v0.4.0 

func (t *TritonSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*TritonSpec) GetStorageUri added in v0.4.0 

func (t *TritonSpec) GetStorageUri() string

func (*TritonSpec) Validate added in v0.4.0 

func (t *TritonSpec) Validate(config *InferenceServicesConfig) error

type VirtualServiceStatus added in v0.3.0 

type VirtualServiceStatus struct {
	URL           string
	CanaryWeight  int
	DefaultWeight int
	// Address holds the information needed for a Route to be the target of an event.
	// +optional
	Address *duckv1beta1.Addressable `json:"address,omitempty"`

	duckv1beta1.Status
}

+k8s:openapi-gen=false 

VirtualServiceStatus captures the status of the virtual service

func (*VirtualServiceStatus) DeepCopy added in v0.3.0 

func (in *VirtualServiceStatus) DeepCopy() *VirtualServiceStatus

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new VirtualServiceStatus.

func (*VirtualServiceStatus) DeepCopyInto added in v0.3.0 

func (in *VirtualServiceStatus) DeepCopyInto(out *VirtualServiceStatus)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

type XGBoostSpec 

type XGBoostSpec struct {
	// The URI of the trained model which contains model.bst
	StorageURI string `json:"storageUri"`
	// Number of thread to be used by XGBoost
	NThread int `json:"nthread,omitempty"`
	// XGBoost KFServer docker image version which defaults to latest release
	RuntimeVersion string `json:"runtimeVersion,omitempty"`
	// Defaults to requests and limits of 1CPU, 2Gb MEM.
	Resources v1.ResourceRequirements `json:"resources,omitempty"`
}

XGBoostSpec defines arguments for configuring XGBoost model serving.

func (*XGBoostSpec) ApplyDefaults 

func (x *XGBoostSpec) ApplyDefaults(config *InferenceServicesConfig)

func (*XGBoostSpec) DeepCopy 

func (in *XGBoostSpec) DeepCopy() *XGBoostSpec

DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new XGBoostSpec.

func (*XGBoostSpec) DeepCopyInto 

func (in *XGBoostSpec) DeepCopyInto(out *XGBoostSpec)

DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.

func (*XGBoostSpec) GetContainer added in v0.3.0 

func (x *XGBoostSpec) GetContainer(modelName string, parallelism int, config *InferenceServicesConfig) *v1.Container

func (*XGBoostSpec) GetResourceRequirements added in v0.3.0 

func (x *XGBoostSpec) GetResourceRequirements() *v1.ResourceRequirements

func (*XGBoostSpec) GetStorageUri 

func (x *XGBoostSpec) GetStorageUri() string

func (*XGBoostSpec) Validate 

func (x *XGBoostSpec) Validate(config *InferenceServicesConfig) error

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.