Class: AWSCDK::Bedrock::CfnAgent::InferenceConfigurationProperty

Inherits:
Jsii::Struct
  • Object
show all
Defined in:
bedrock/cfn_agent.rb

Overview

Base inference parameters to pass to a model in a call to Converse or ConverseStream . For more information, see Inference parameters for foundation models .

If you need to pass additional parameters that the model supports, use the additional_model_request_fields request field in the call to Converse or ConverseStream . For more information, see Model parameters .

See Also:

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(maximum_length: nil, stop_sequences: nil, temperature: nil, top_k: nil, top_p: nil) ⇒ InferenceConfigurationProperty

Returns a new instance of InferenceConfigurationProperty.

Parameters:

  • maximum_length (Numeric, nil) (defaults to: nil) —

    The maximum number of tokens allowed in the generated response.

  • stop_sequences (Array<String>, nil) (defaults to: nil) —

    A list of stop sequences.

  • temperature (Numeric, nil) (defaults to: nil) —

    The likelihood of the model selecting higher-probability options while generating a response.

  • top_k (Numeric, nil) (defaults to: nil) —

    While generating a response, the model determines the probability of the following token at each point of generation.

  • top_p (Numeric, nil) (defaults to: nil) —

    The percentage of most-likely candidates that the model considers for the next token.



1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
 
# File 'bedrock/cfn_agent.rb', line 1412

def initialize(maximum_length: nil, stop_sequences: nil, temperature: nil, top_k: nil, top_p: nil)
  @maximum_length = maximum_length
  Jsii::Type.check_type(@maximum_length, JSON.parse(Base64.strict_decode64("eyJwcmltaXRpdmUiOiJudW1iZXIifQ==")), "maximumLength") unless @maximum_length.nil?
  @stop_sequences = stop_sequences
  Jsii::Type.check_type(@stop_sequences, JSON.parse(Base64.strict_decode64("eyJjb2xsZWN0aW9uIjp7ImVsZW1lbnR0eXBlIjp7InByaW1pdGl2ZSI6InN0cmluZyJ9LCJraW5kIjoiYXJyYXkifX0=")), "stopSequences") unless @stop_sequences.nil?
  @temperature = temperature
  Jsii::Type.check_type(@temperature, JSON.parse(Base64.strict_decode64("eyJwcmltaXRpdmUiOiJudW1iZXIifQ==")), "temperature") unless @temperature.nil?
  @top_k = top_k
  Jsii::Type.check_type(@top_k, JSON.parse(Base64.strict_decode64("eyJwcmltaXRpdmUiOiJudW1iZXIifQ==")), "topK") unless @top_k.nil?
  @top_p = top_p
  Jsii::Type.check_type(@top_p, JSON.parse(Base64.strict_decode64("eyJwcmltaXRpdmUiOiJudW1iZXIifQ==")), "topP") unless @top_p.nil?
end

Instance Attribute Details

#maximum_length ⇒ Numeric? (readonly)

The maximum number of tokens allowed in the generated response.

Returns:

  • (Numeric, nil)

See Also:



1429
1430
1431
 
# File 'bedrock/cfn_agent.rb', line 1429

def maximum_length
  @maximum_length
end

#stop_sequences ⇒ Array<String>? (readonly)

A list of stop sequences.

A stop sequence is a sequence of characters that causes the model to stop generating the response.

Returns:

  • (Array<String>, nil)

See Also:



1436
1437
1438
 
# File 'bedrock/cfn_agent.rb', line 1436

def stop_sequences
  @stop_sequences
end

#temperature ⇒ Numeric? (readonly)

The likelihood of the model selecting higher-probability options while generating a response.

A lower value makes the model more likely to choose higher-probability options, while a higher value makes the model more likely to choose lower-probability options.

The default value is the default value for the model that you are using. For more information, see Inference parameters for foundation models .

Returns:

  • (Numeric, nil)

See Also:



1445
1446
1447
 
# File 'bedrock/cfn_agent.rb', line 1445

def temperature
  @temperature
end

#top_k ⇒ Numeric? (readonly)

While generating a response, the model determines the probability of the following token at each point of generation.

The value that you set for top_k is the number of most-likely candidates from which the model chooses the next token in the sequence. For example, if you set top_k to 50, the model selects the next token from among the top 50 most likely choices.

Returns:

  • (Numeric, nil)

See Also:



1452
1453
1454
 
# File 'bedrock/cfn_agent.rb', line 1452

def top_k
  @top_k
end

#top_p ⇒ Numeric? (readonly)

The percentage of most-likely candidates that the model considers for the next token.

For example, if you choose a value of 0.8 for top_p , the model selects from the top 80% of the probability distribution of tokens that could be next in the sequence.

The default value is the default value for the model that you are using. For more information, see Inference parameters for foundation models .

Returns:

  • (Numeric, nil)

See Also:



1461
1462
1463
 
# File 'bedrock/cfn_agent.rb', line 1461

def top_p
  @top_p
end

Class Method Details

.jsii_properties ⇒ Object 



1463
1464
1465
1466
1467
1468
1469
1470
1471
 
# File 'bedrock/cfn_agent.rb', line 1463

def self.jsii_properties
  {
    :maximum_length => "maximumLength",
    :stop_sequences => "stopSequences",
    :temperature => "temperature",
    :top_k => "topK",
    :top_p => "topP",
  }
end

Instance Method Details

#to_jsii ⇒ Object 



1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
 
# File 'bedrock/cfn_agent.rb', line 1473

def to_jsii
  result = {}
  result.merge!({
    "maximumLength" => @maximum_length,
    "stopSequences" => @stop_sequences,
    "temperature" => @temperature,
    "topK" => @top_k,
    "topP" => @top_p,
  })
  result.compact
end