View Source GoogleApi.AIPlatform.V1.Model.GoogleCloudAiplatformV1SchemaTextPromptDatasetMetadata (google_api_ai_platform v0.24.0)

The metadata of Datasets that contain Text Prompt data.

Attributes

  • candidateCount (type: String.t, default: nil) - Number of candidates.
  • gcsUri (type: String.t, default: nil) - The Google Cloud Storage URI that stores the prompt data.
  • groundingConfig (type: GoogleApi.AIPlatform.V1.Model.GoogleCloudAiplatformV1SchemaPredictParamsGroundingConfig.t, default: nil) - Grounding checking configuration.
  • hasPromptVariable (type: boolean(), default: nil) - Whether the prompt dataset has prompt variable.
  • logprobs (type: boolean(), default: nil) - Whether or not the user has enabled logit probabilities in the model parameters.
  • maxOutputTokens (type: String.t, default: nil) - Value of the maximum number of tokens generated set when the dataset was saved.
  • note (type: String.t, default: nil) - User-created prompt note. Note size limit is 2KB.
  • promptApiSchema (type: GoogleApi.AIPlatform.V1.Model.GoogleCloudAiplatformV1SchemaPromptApiSchema.t, default: nil) - The API schema of the prompt to support both UI and SDK usages.
  • promptType (type: String.t, default: nil) - Type of the prompt dataset.
  • seedEnabled (type: boolean(), default: nil) - Seeding enables model to return a deterministic response on a best effort basis. Determinism isn't guaranteed. This field determines whether or not seeding is enabled.
  • seedValue (type: String.t, default: nil) - The actual value of the seed.
  • stopSequences (type: list(String.t), default: nil) - Customized stop sequences.
  • systemInstruction (type: String.t, default: nil) - The content of the prompt dataset system instruction.
  • systemInstructionGcsUri (type: String.t, default: nil) - The Google Cloud Storage URI that stores the system instruction, starting with gs://.
  • temperature (type: number(), default: nil) - Temperature value used for sampling set when the dataset was saved. This value is used to tune the degree of randomness.
  • text (type: String.t, default: nil) - The content of the prompt dataset.
  • topK (type: String.t, default: nil) - Top K value set when the dataset was saved. This value determines how many candidates with highest probability from the vocab would be selected for each decoding step.
  • topP (type: number(), default: nil) - Top P value set when the dataset was saved. Given topK tokens for decoding, top candidates will be selected until the sum of their probabilities is topP.

Summary

Types

t()

Functions

decode(value, options)

Unwrap a decoded JSON object into its complex fields.

Types

t()

@type t() ::
  %GoogleApi.AIPlatform.V1.Model.GoogleCloudAiplatformV1SchemaTextPromptDatasetMetadata{
    candidateCount: String.t() | nil,
    gcsUri: String.t() | nil,
    groundingConfig:
      GoogleApi.AIPlatform.V1.Model.GoogleCloudAiplatformV1SchemaPredictParamsGroundingConfig.t()
      | nil,
    hasPromptVariable: boolean() | nil,
    logprobs: boolean() | nil,
    maxOutputTokens: String.t() | nil,
    note: String.t() | nil,
    promptApiSchema:
      GoogleApi.AIPlatform.V1.Model.GoogleCloudAiplatformV1SchemaPromptApiSchema.t()
      | nil,
    promptType: String.t() | nil,
    seedEnabled: boolean() | nil,
    seedValue: String.t() | nil,
    stopSequences: [String.t()] | nil,
    systemInstruction: String.t() | nil,
    systemInstructionGcsUri: String.t() | nil,
    temperature: number() | nil,
    text: String.t() | nil,
    topK: String.t() | nil,
    topP: number() | nil
  }

Functions

decode(value, options)

@spec decode(struct(), keyword()) :: struct()

Unwrap a decoded JSON object into its complex fields.