Bundle Google Gemini support for tuned models (-lib gemini)
Tuner
Fine-tuning model operations
Inherits: OAuthEndPoint
Operations
Create # function
Creates a tuned model
function : Create(display_name:String, base_model:String, tuning_task:TuningTask) ~ API.Google.Gemini.Tuning.TunedModelParameters
| Name | Type | Description |
|---|---|---|
| display_name | String | the name to display for this model in user interfaces |
| base_model | String | the name of the base model to tune from |
| tuning_task | TuningTask | the tuning task that creates the tuned model |
Return
| Type | Description |
|---|---|
| TunedModel | snapshot of tuned model and metadata |
Create # function
Creates a tuned model
function : Create(display_name:String, description:String, tuning_task:TuningTask, tune_model_source:Pair<String,String>, base_model:String, temperature:Float, top_p:Float, top_k:Int) ~ API.Google.Gemini.Tuning.TunedModelParameters
| Name | Type | Description |
|---|---|---|
| display_name | String | the name to display for this model in user interfaces |
| description | String | a short description of this model |
| tuning_task | TuningTask | the tuning task that creates the tuned model |
| tune_model_source | Pair<String,String> | used as the starting point for training the new model |
| base_model | String | the name of the base model to tune from |
| temperature | Float | controls the randomness of the output |
| top_p | Float | for nucleus sampling considers the smallest set of tokens whose probability sum is at least top-p. |
| top_k | Int | for top-k sampling considers the set of top-k most probable tokens. this value specifies default to be used by the backend while making the call to the model |
Return
| Type | Description |
|---|---|
| TunedModel | snapshot of tuned model and metadata |
DeleteAllModels # function
Deletes all trained models
function : DeleteAllModels() ~ BoolReturn
| Type | Description |
|---|---|
| Bool | true if successful, false otherwise |
Get # function
Creates a tuned model
function : Get(model_id:String) ~ TunedModelParameters
| Name | Type | Description |
|---|---|---|
| model_id | String | resource model identifier |
Return
| Type | Description |
|---|---|
| TunedModel | tuned model |
List # function
Lists tuned models owned by the user
function : List() ~ Vector<TunedModel>Return
| Type | Description |
|---|---|
| Vector<TunedModel> | list of tuned models |
ToJson #
JSON representation of the instance
method : public : ToJson() ~ JsonElementReturn
| Type | Description |
|---|---|
| JsonElement | JSON representation of the instance |