inferences

Documentation

Index

• func CreateInferenceDeployment(c *gcorecloud.ServiceClient, opts CreateInferenceDeploymentOptsBuilder) (r tasks.Result)
• func DeleteInferenceDeployment(c *gcorecloud.ServiceClient, name string) (r tasks.Result)
• func UpdateInferenceDeployment(c *gcorecloud.ServiceClient, name string, ...) (r tasks.Result)
• type Container
• type ContainerDeployStatus
• type ContainerScale
• type ContainerScaleTrigger
• type CreateContainerOpts
• type CreateInferenceDeploymentOpts
  • func (opts CreateInferenceDeploymentOpts) ToRegistryCredentialCreateMap() (map[string]interface{}, error)
• type CreateInferenceDeploymentOptsBuilder
• type CreateLoggingOpts
• type ExecProbe
• type GetResult
  • func GetInferenceDeployment(c *gcorecloud.ServiceClient, name string) (r GetResult)
  • func (r GetResult) Extract() (*InferenceDeployment, error)
  • func (r GetResult) ExtractInto(v interface{}) error
• type HttpGetProbe
• type InferenceDeployment
  • func ListAllInferenceDeployments(c *gcorecloud.ServiceClient) ([]InferenceDeployment, error)
• type ListResult
  • func (r ListResult) Extract() ([]InferenceDeployment, error)
  • func (r ListResult) ExtractInto(v interface{}) error
• type Logging
• type LoggingRetentionPolicy
• type Probe
• type ProbeConfiguration
• type Probes
• type ScaleTriggerHttp
• type ScaleTriggerSqs
• type ScaleTriggerThreshold
• type TcpSocketProbe
• type UpdateInferenceDeploymentOpts
  • func (opts UpdateInferenceDeploymentOpts) ToRegistryCredentialUpdateMap() (map[string]interface{}, error)
• type UpdateInferenceDeploymentOptsBuilder

Functions

func CreateInferenceDeployment

func CreateInferenceDeployment(c *gcorecloud.ServiceClient, opts CreateInferenceDeploymentOptsBuilder) (r tasks.Result)

CreateInferenceDeployment create FaaS function.

func DeleteInferenceDeployment

func DeleteInferenceDeployment(c *gcorecloud.ServiceClient, name string) (r tasks.Result)

DeleteInferenceDeployment accepts a unique ID and deletes the inference deployment associated with it.

func UpdateInferenceDeployment

func UpdateInferenceDeployment(c *gcorecloud.ServiceClient, name string, opts UpdateInferenceDeploymentOptsBuilder) (r tasks.Result)

UpdateInferenceDeployment update existing inference deployment.

Types

type Container

type Container struct {
	RegionID     int                   `json:"region_id"`
	Address      string                `json:"address"`
	Scale        ContainerScale        `json:"scale"`
	DeployStatus ContainerDeployStatus `json:"deploy_status"`
	ErrorMessage string                `json:"error_message"`
}

type ContainerDeployStatus

type ContainerDeployStatus struct {
	Total int `json:"total"`
	Ready int `json:"ready"`
}

type ContainerScale

type ContainerScale struct {
	Min             int                   `json:"min"`
	Max             int                   `json:"max"`
	CooldownPeriod  *int                  `json:"cooldown_period"`
	PollingInterval *int                  `json:"polling_interval"`
	Triggers        ContainerScaleTrigger `json:"triggers"`
}

type ContainerScaleTrigger

type ContainerScaleTrigger struct {
	Cpu            *ScaleTriggerThreshold `json:"cpu,omitempty"`
	GpuMemory      *ScaleTriggerThreshold `json:"gpu_memory,omitempty"`
	GpuUtilization *ScaleTriggerThreshold `json:"gpu_utilization,omitempty"`
	Memory         *ScaleTriggerThreshold `json:"memory,omitempty"`
	Http           *ScaleTriggerHttp      `json:"http,omitempty"`
	Sqs            *ScaleTriggerSqs       `json:"sqs,omitempty"`
}

type CreateContainerOpts

type CreateContainerOpts struct {
	RegionID int            `json:"region_id"`
	Scale    ContainerScale `json:"scale"`
}

CreateContainerOpts represents options used to create a container.

type CreateInferenceDeploymentOpts

type CreateInferenceDeploymentOpts struct {
	Name            string                `json:"name"`
	Image           string                `json:"image"`
	ListeningPort   int                   `json:"listening_port"`
	Description     string                `json:"description"`
	AuthEnabled     bool                  `json:"auth_enabled"`
	Containers      []CreateContainerOpts `json:"containers"`
	Timeout         *int                  `json:"timeout,omitempty"`
	Envs            map[string]string     `json:"envs,omitempty"`
	Command         []string              `json:"command,omitempty"`
	CredentialsName *string               `json:"credentials_name,omitempty"`
	Logging         *CreateLoggingOpts    `json:"logging,omitempty"`
	Probes          *Probes               `json:"probes,omitempty"`
	FlavorName      string                `json:"flavor_name"`
}

CreateInferenceDeploymentOpts represents options used to create a function.

func (CreateInferenceDeploymentOpts) ToRegistryCredentialCreateMap

func (opts CreateInferenceDeploymentOpts) ToRegistryCredentialCreateMap() (map[string]interface{}, error)

ToInferenceInstanceCreateMap builds a request body from CreateInferenceDeploymentOpts.

type CreateInferenceDeploymentOptsBuilder

type CreateInferenceDeploymentOptsBuilder interface {
	ToRegistryCredentialCreateMap() (map[string]interface{}, error)
}

CreateInferenceDeploymentOptsBuilder allows extensions to add additional parameters to the request.

type CreateLoggingOpts

type CreateLoggingOpts struct {
	Enabled             bool                   `json:"enabled"`
	DestinationRegionID int                    `json:"destination_region_id"`
	TopicName           string                 `json:"topic_name"`
	RetentionPolicy     LoggingRetentionPolicy `json:"retention_policy"`
}

CreateLoggingOpts represents options used to create a logging.

type ExecProbe

type ExecProbe struct {
	Command []string `json:"command"`
}

type GetResult

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

GetResult represents the result of a get operation. Call its Extract method to interpret it as a InferenceDeployment.

func GetInferenceDeployment

func GetInferenceDeployment(c *gcorecloud.ServiceClient, name string) (r GetResult)

GetInferenceDeployment get inference deployment instance.

func (GetResult) Extract

func (r GetResult) Extract() (*InferenceDeployment, error)

Extract is a function that accepts a result and extracts a inference resource.

func (GetResult) ExtractInto

func (r GetResult) ExtractInto(v interface{}) error

type HttpGetProbe

type HttpGetProbe struct {
	Headers map[string]string `json:"headers"`
	Host    *string           `json:"host"`
	Path    string            `json:"path"`
	Port    int               `json:"port"`
	Schema  string            `json:"schema"`
}

type InferenceDeployment

type InferenceDeployment struct {
	ProjectID       int               `json:"project_id"`
	Name            string            `json:"name"`
	Description     string            `json:"description"`
	Image           string            `json:"image"`
	ListeningPort   int               `json:"listening_port"`
	CreatedAt       *string           `json:"created_at"`
	AuthEnabled     bool              `json:"auth_enabled"`
	Address         string            `json:"address"`
	Timeout         int               `json:"timeout"`
	Envs            map[string]string `json:"envs"`
	FlavorName      string            `json:"flavor_name"`
	Command         *string           `json:"command"`
	CredentialsName string            `json:"credentials_name"`
	Logging         *Logging          `json:"logging"`
	Probes          *Probes           `json:"probes"`
	Containers      []Container       `json:"containers"`
	Status          string            `json:"status"`
}

func ListAllInferenceDeployments

func ListAllInferenceDeployments(c *gcorecloud.ServiceClient) ([]InferenceDeployment, error)

ListAllInferenceDeployments lists all inference deployments.

type ListResult

type ListResult struct {
	gcorecloud.Result
}

func (ListResult) Extract

func (r ListResult) Extract() ([]InferenceDeployment, error)

Extract is a function that accepts a result and extracts a inference deployment resource.

func (ListResult) ExtractInto

func (r ListResult) ExtractInto(v interface{}) error

type Logging

type Logging struct {
	Enabled                  bool    `json:"enabled"`
	DestinationRegionID      *int    `json:"destination_region_id"`
	TopicName                *string `json:"topic_name"`
	RetentionPolicy          *int    `json:"retention_policy"`
	OpensearchDashboardsLink string  `json:"opensearch_dashboards_link"`
}

type LoggingRetentionPolicy

type LoggingRetentionPolicy struct {
	Period *int `json:"period"`
}

type Probe

type Probe struct {
	FailureThreshold    int             `json:"failure_threshold"`
	InitialDelaySeconds int             `json:"initial_delay_seconds"`
	PeriodSeconds       int             `json:"period_seconds"`
	TimeoutSeconds      int             `json:"timeout_seconds"`
	SuccessThreshold    int             `json:"success_threshold"`
	Exec                *ExecProbe      `json:"exec"`
	TcpSocket           *TcpSocketProbe `json:"tcp_socket"`
	HttpGet             *HttpGetProbe   `json:"http_get"`
}

type ProbeConfiguration

type ProbeConfiguration struct {
	Enabled bool   `json:"enabled"`
	Probe   *Probe `json:"probe"`
}

type Probes

type Probes struct {
	LivenessProbe  *ProbeConfiguration `json:"liveness_probe"`
	ReadinessProbe *ProbeConfiguration `json:"readiness_probe"`
	StartupProbe   *ProbeConfiguration `json:"startup_probe"`
}

type ScaleTriggerHttp

type ScaleTriggerHttp struct {
	Rate   *int `json:"rate"`
	Window *int `json:"window"`
}

type ScaleTriggerSqs

type ScaleTriggerSqs struct {
	QueueURL              string  `json:"queue_url"`
	QueueLength           int     `json:"queue_length"`
	ActivationQueueLength int     `json:"activation_queue_length"`
	ScaleOnFlight         bool    `json:"scale_on_flight"`
	ScaleOnDelayed        bool    `json:"scale_on_delayed"`
	AwsRegion             string  `json:"aws_region"`
	AwsEndpoint           *string `json:"aws_endpoint"`
	SecretName            string  `json:"secret_name"`
}

type ScaleTriggerThreshold

type ScaleTriggerThreshold struct {
	Threshold int `json:"threshold"`
}

type TcpSocketProbe

type TcpSocketProbe struct {
	Port int `json:"port"`
}

type UpdateInferenceDeploymentOpts

type UpdateInferenceDeploymentOpts struct {
	Description     *string               `json:"description,omitempty"`
	Image           *string               `json:"image,omitempty"`
	ListeningPort   *int                  `json:"listening_port,omitempty"`
	AuthEnabled     *bool                 `json:"auth_enabled,omitempty"`
	Containers      []CreateContainerOpts `json:"containers,omitempty"`
	Timeout         *int                  `json:"timeout"`
	Envs            map[string]string     `json:"envs,omitempty"`
	Command         []string              `json:"command"`
	Logging         *CreateLoggingOpts    `json:"logging,omitempty"`
	Probes          *Probes               `json:"probes,omitempty"`
	FlavorName      *string               `json:"flavor_name"`
	CredentialsName *string               `json:"credentials_name"`
}

UpdateInferenceDeploymentOpts represents options used to update a function.

func (UpdateInferenceDeploymentOpts) ToRegistryCredentialUpdateMap

func (opts UpdateInferenceDeploymentOpts) ToRegistryCredentialUpdateMap() (map[string]interface{}, error)

ToInferenceDeploymentUpdateMap builds a request body from UpdateInferenceDeploymentOpts.

type UpdateInferenceDeploymentOptsBuilder

type UpdateInferenceDeploymentOptsBuilder interface {
	ToRegistryCredentialUpdateMap() (map[string]interface{}, error)
}

UpdateInferenceDeploymentOptsBuilder allows extensions to add additional parameters to the request.