REST Resource: projects.conversationModels

Resource: ConversationModel

Represents a conversation model.

JSON representation 
{
  "name": string,
  "displayName": string,
  "createTime": string,
  "datasets": [
    {
      object (InputDataset)
    }
  ],
  "state": enum (State),
  "languageCode": string,

  // Union field model_metadata can be only one of the following:
  "articleSuggestionModelMetadata": {
    object (ArticleSuggestionModelMetadata)
  },
  "smartReplyModelMetadata": {
    object (SmartReplyModelMetadata)
  }
  // End of list of possible types for union field model_metadata.
}
Fields
name

string

ConversationModel resource name. Format: projects/<Project ID>/conversationModels/<Conversation Model ID>
displayName

string

Required. The display name of the model. At most 64 bytes long.
createTime

string (Timestamp format)

Output only. Creation time of this model.

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".
datasets[]

object (InputDataset)

Required. Datasets used to create model.
state

enum (State)

Output only. State of the model. A model can only serve prediction requests after it gets deployed.
languageCode

string

Language code for the conversation model. If not specified, the language is en-US. Language at ConversationModel should be set for all non en-us languages. This should be a BCP-47 language tag. Example: "en-US".
Union field model_metadata. Required. The model metadata that is specific to the problem type. Must match the metadata type of the dataset used to train the model. model_metadata can be only one of the following:
articleSuggestionModelMetadata

object (ArticleSuggestionModelMetadata)

Metadata for article suggestion models.
smartReplyModelMetadata

object (SmartReplyModelMetadata)

Metadata for smart reply models.

State

State of the model.

Enums
STATE_UNSPECIFIED Should not be used, an un-set enum has this value by default.
CREATING Model being created.
UNDEPLOYED Model is not deployed but ready to deploy.
DEPLOYING Model is deploying.
DEPLOYED Model is deployed and ready to use.
UNDEPLOYING Model is undeploying.
DELETING Model is deleting.
FAILED Model is in error state. Not ready to deploy and use.
PENDING Model is being created but the training has not started, The model may remain in this state until there is enough capacity to start training.

ArticleSuggestionModelMetadata

Metadata for article suggestion models.

JSON representation 
{
  "trainingModelType": enum (ModelType)
}
Fields
trainingModelType

enum (ModelType)

Optional. Type of the article suggestion model. If not provided, modelType is used.

ModelType

Model type.

Enums
MODEL_TYPE_UNSPECIFIED ModelType unspecified.
SMART_REPLY_DUAL_ENCODER_MODEL ModelType smart reply dual encoder model.
SMART_REPLY_BERT_MODEL ModelType smart reply bert model.

SmartReplyModelMetadata

Metadata for smart reply models.

JSON representation 
{
  "trainingModelType": enum (ModelType)
}
Fields
trainingModelType

enum (ModelType)

Optional. Type of the smart reply model. If not provided, modelType is used.

Methods

create

 Creates a model.

delete

 Deletes a model.

deploy

 Deploys a model.

get

 Gets conversation model.

list

 Lists conversation models.

undeploy

 Undeploys a model.