Class: Google::Apis::AiplatformV1beta1::GoogleCloudAiplatformV1beta1ReinforcementTuningExample

Inherits:
Object
  • Object
show all
Includes:
Core::Hashable, Core::JsonObjectSupport
Defined in:
lib/google/apis/aiplatform_v1beta1/classes.rb,
lib/google/apis/aiplatform_v1beta1/representations.rb,
lib/google/apis/aiplatform_v1beta1/representations.rb

Overview

User-facing format for Gemini Reinforcement Tuning examples on Vertex.

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(**args) ⇒ GoogleCloudAiplatformV1beta1ReinforcementTuningExample

Returns a new instance of GoogleCloudAiplatformV1beta1ReinforcementTuningExample.



49488
49489
49490
 
# File 'lib/google/apis/aiplatform_v1beta1/classes.rb', line 49488

def initialize(**args)
   update!(**args)
end

Instance Attribute Details

#contents ⇒ Array<Google::Apis::AiplatformV1beta1::GoogleCloudAiplatformV1beta1Content>

Multi-turn contents that represents the Prompt. Corresponds to the JSON property contents

Returns:



49458
49459
49460
 
# File 'lib/google/apis/aiplatform_v1beta1/classes.rb', line 49458

def contents
  @contents
end

#references ⇒ Hash<String,String>

References for the given prompt. The key is the name of the reference, and the value is the reference itself. Users can use this field together with the reward configurations to calculate rewards for reinforcement tuning. For example, users can set the following references: ` "concise_answer": "Yes", "verbose_answer": "The answer is Yes" ` Then in a ReinforcementTuningCodeExecutionRewardScorer reward function config, for example, they can define a python code snippet as follows: def evaluate( example, response) -> float: response_str = response.get("parts", [])0 references = example.get("references", ``) if response_str == references.get(" concise_answer"): return 1.0 return -1.0 In this case, references can serve the purpose of holding the ground truth of this example in the training/ validation dataset. Corresponds to the JSON property references

Returns:

  • (Hash<String,String>) 


49474
49475
49476
 
# File 'lib/google/apis/aiplatform_v1beta1/classes.rb', line 49474

def references
  @references
end

#system_instruction ⇒ Google::Apis::AiplatformV1beta1::GoogleCloudAiplatformV1beta1Content

The structured data content of a message. A Content message contains a role field, which indicates the producer of the content, and a parts field, which contains the multi-part data of the message. Corresponds to the JSON property systemInstruction

Returns:



49481
49482
49483
 
# File 'lib/google/apis/aiplatform_v1beta1/classes.rb', line 49481

def system_instruction
  @system_instruction
end

#tools ⇒ Array<Google::Apis::AiplatformV1beta1::GoogleCloudAiplatformV1beta1Tool>

Optional. Corresponds to tools in user-facing GenerateContentRequest. Corresponds to the JSON property tools

Returns:



49486
49487
49488
 
# File 'lib/google/apis/aiplatform_v1beta1/classes.rb', line 49486

def tools
  @tools
end

Instance Method Details

#update!(**args) ⇒ Object

Update properties of this object



49493
49494
49495
49496
49497
49498
 
# File 'lib/google/apis/aiplatform_v1beta1/classes.rb', line 49493

def update!(**args)
  @contents = args[:contents] if args.key?(:contents)
  @references = args[:references] if args.key?(:references)
  @system_instruction = args[:system_instruction] if args.key?(:system_instruction)
  @tools = args[:tools] if args.key?(:tools)
end