Class DistillationSpec
-
- All Implemented Interfaces:
public abstract class DistillationSpec extends JsonSerializable
Distillation tuning spec for tuning.
-
-
Nested Class Summary
Nested Classes Modifier and Type Class Description public abstract classDistillationSpec.BuilderBuilder for DistillationSpec.
-
Field Summary
Fields Modifier and Type Field Description public final static StringMAX_READ_LENGTH_PROPERTY
-
Constructor Summary
Constructors Constructor Description DistillationSpec()
-
Method Summary
Modifier and Type Method Description abstract Optional<String>promptDatasetUri()The GCS URI of the prompt dataset to use during distillation. abstract Optional<String>baseTeacherModel()The base teacher model that is being distilled. abstract Optional<DistillationHyperParameters>hyperParameters()Optional. abstract Optional<String>pipelineRootDirectory()Deprecated. abstract Optional<String>studentModel()The student model that is being tuned, e.g., "google/gemma-2b-1.1-it". abstract Optional<String>trainingDatasetUri()Deprecated. abstract Optional<String>tunedTeacherModelSource()The resource name of the Tuned teacher model. abstract Optional<String>validationDatasetUri()Optional. abstract Optional<TuningMode>tuningMode()Tuning mode for tuning. static DistillationSpec.Builderbuilder()Instantiates a builder for DistillationSpec. abstract DistillationSpec.BuildertoBuilder()Creates a builder with the same values as this instance. static DistillationSpecfromJson(String jsonString)Deserializes a JSON string to a DistillationSpec object. -
Methods inherited from class com.google.genai.JsonSerializable
fromJsonNode, fromJsonString, objectMapper, setMaxReadLength, stringToJsonNode, toJson, toJsonNode, toJsonString -
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
-
-
Method Detail
-
promptDatasetUri
abstract Optional<String> promptDatasetUri()
The GCS URI of the prompt dataset to use during distillation.
-
baseTeacherModel
abstract Optional<String> baseTeacherModel()
The base teacher model that is being distilled. See [Supported models](https://cloud.google.com/vertex-ai/generative-ai/docs/model-reference/tuning#supported_models).
-
hyperParameters
abstract Optional<DistillationHyperParameters> hyperParameters()
Optional. Hyperparameters for Distillation.
-
pipelineRootDirectory
abstract Optional<String> pipelineRootDirectory()
Deprecated. A path in a Cloud Storage bucket, which will be treated as the root output directory of the distillation pipeline. It is used by the system to generate the paths of output artifacts.
-
studentModel
abstract Optional<String> studentModel()
The student model that is being tuned, e.g., "google/gemma-2b-1.1-it". Deprecated. Use base_model instead.
-
trainingDatasetUri
abstract Optional<String> trainingDatasetUri()
Deprecated. Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file.
-
tunedTeacherModelSource
abstract Optional<String> tunedTeacherModelSource()
The resource name of the Tuned teacher model. Format: `projects/{project}/locations/{location}/models/{model}`.
-
validationDatasetUri
abstract Optional<String> validationDatasetUri()
Optional. Cloud Storage path to file containing validation dataset for tuning. The dataset must be formatted as a JSONL file.
-
tuningMode
abstract Optional<TuningMode> tuningMode()
Tuning mode for tuning.
-
builder
static DistillationSpec.Builder builder()
Instantiates a builder for DistillationSpec.
-
toBuilder
abstract DistillationSpec.Builder toBuilder()
Creates a builder with the same values as this instance.
-
fromJson
static DistillationSpec fromJson(String jsonString)
Deserializes a JSON string to a DistillationSpec object.
-
-
-
-