Class: OCI::GenerativeAiInference::Models::LlamaLlmInferenceRequest

Inherits:
LlmInferenceRequest show all
Defined in:
lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb

Overview

Details for the text generation request for Llama models.

Constant Summary

Constants inherited from LlmInferenceRequest

OCI::GenerativeAiInference::Models::LlmInferenceRequest::RUNTIME_TYPE_ENUM

Instance Attribute Summary collapse

Attributes inherited from LlmInferenceRequest

#runtime_type

Class Method Summary collapse

Instance Method Summary collapse

Methods inherited from LlmInferenceRequest

get_subtype

Constructor Details

#initialize(attributes = {}) ⇒ LlamaLlmInferenceRequest

Initializes the object

Parameters:

  • attributes (Hash) (defaults to: {}) —

    Model attributes in the form of hash

Options Hash (attributes):

  • :prompt (String) —

    The value to assign to the #prompt property

  • :is_stream (BOOLEAN) —

    The value to assign to the #is_stream property

  • :num_generations (Integer) —

    The value to assign to the #num_generations property

  • :is_echo (BOOLEAN) —

    The value to assign to the #is_echo property

  • :top_k (Integer) —

    The value to assign to the #top_k property

  • :top_p (Float) —

    The value to assign to the #top_p property

  • :temperature (Float) —

    The value to assign to the #temperature property

  • :frequency_penalty (Float) —

    The value to assign to the #frequency_penalty property

  • :presence_penalty (Float) —

    The value to assign to the #presence_penalty property

  • :stop (Array<String>) —

    The value to assign to the #stop property

  • :log_probs (Integer) —

    The value to assign to the #log_probs property

  • :max_tokens (Integer) —

    The value to assign to the #max_tokens property



135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 135

def initialize(attributes = {})
  return unless attributes.is_a?(Hash)

  attributes['runtimeType'] = 'LLAMA'

  super(attributes)

  # convert string to symbol for hash key
  attributes = attributes.each_with_object({}) { |(k, v), h| h[k.to_sym] = v }

  self.prompt = attributes[:'prompt'] if attributes[:'prompt']

  self.is_stream = attributes[:'isStream'] unless attributes[:'isStream'].nil?
  self.is_stream = false if is_stream.nil? && !attributes.key?(:'isStream') # rubocop:disable Style/StringLiterals

  raise 'You cannot provide both :isStream and :is_stream' if attributes.key?(:'isStream') && attributes.key?(:'is_stream')

  self.is_stream = attributes[:'is_stream'] unless attributes[:'is_stream'].nil?
  self.is_stream = false if is_stream.nil? && !attributes.key?(:'isStream') && !attributes.key?(:'is_stream') # rubocop:disable Style/StringLiterals

  self.num_generations = attributes[:'numGenerations'] if attributes[:'numGenerations']

  raise 'You cannot provide both :numGenerations and :num_generations' if attributes.key?(:'numGenerations') && attributes.key?(:'num_generations')

  self.num_generations = attributes[:'num_generations'] if attributes[:'num_generations']

  self.is_echo = attributes[:'isEcho'] unless attributes[:'isEcho'].nil?
  self.is_echo = false if is_echo.nil? && !attributes.key?(:'isEcho') # rubocop:disable Style/StringLiterals

  raise 'You cannot provide both :isEcho and :is_echo' if attributes.key?(:'isEcho') && attributes.key?(:'is_echo')

  self.is_echo = attributes[:'is_echo'] unless attributes[:'is_echo'].nil?
  self.is_echo = false if is_echo.nil? && !attributes.key?(:'isEcho') && !attributes.key?(:'is_echo') # rubocop:disable Style/StringLiterals

  self.top_k = attributes[:'topK'] if attributes[:'topK']

  raise 'You cannot provide both :topK and :top_k' if attributes.key?(:'topK') && attributes.key?(:'top_k')

  self.top_k = attributes[:'top_k'] if attributes[:'top_k']

  self.top_p = attributes[:'topP'] if attributes[:'topP']
  self.top_p = 1.0 if top_p.nil? && !attributes.key?(:'topP') # rubocop:disable Style/StringLiterals

  raise 'You cannot provide both :topP and :top_p' if attributes.key?(:'topP') && attributes.key?(:'top_p')

  self.top_p = attributes[:'top_p'] if attributes[:'top_p']
  self.top_p = 1.0 if top_p.nil? && !attributes.key?(:'topP') && !attributes.key?(:'top_p') # rubocop:disable Style/StringLiterals

  self.temperature = attributes[:'temperature'] if attributes[:'temperature']
  self.temperature = 1.0 if temperature.nil? && !attributes.key?(:'temperature') # rubocop:disable Style/StringLiterals

  self.frequency_penalty = attributes[:'frequencyPenalty'] if attributes[:'frequencyPenalty']
  self.frequency_penalty = 0.0 if frequency_penalty.nil? && !attributes.key?(:'frequencyPenalty') # rubocop:disable Style/StringLiterals

  raise 'You cannot provide both :frequencyPenalty and :frequency_penalty' if attributes.key?(:'frequencyPenalty') && attributes.key?(:'frequency_penalty')

  self.frequency_penalty = attributes[:'frequency_penalty'] if attributes[:'frequency_penalty']
  self.frequency_penalty = 0.0 if frequency_penalty.nil? && !attributes.key?(:'frequencyPenalty') && !attributes.key?(:'frequency_penalty') # rubocop:disable Style/StringLiterals

  self.presence_penalty = attributes[:'presencePenalty'] if attributes[:'presencePenalty']
  self.presence_penalty = 0.0 if presence_penalty.nil? && !attributes.key?(:'presencePenalty') # rubocop:disable Style/StringLiterals

  raise 'You cannot provide both :presencePenalty and :presence_penalty' if attributes.key?(:'presencePenalty') && attributes.key?(:'presence_penalty')

  self.presence_penalty = attributes[:'presence_penalty'] if attributes[:'presence_penalty']
  self.presence_penalty = 0.0 if presence_penalty.nil? && !attributes.key?(:'presencePenalty') && !attributes.key?(:'presence_penalty') # rubocop:disable Style/StringLiterals

  self.stop = attributes[:'stop'] if attributes[:'stop']

  self.log_probs = attributes[:'logProbs'] if attributes[:'logProbs']

  raise 'You cannot provide both :logProbs and :log_probs' if attributes.key?(:'logProbs') && attributes.key?(:'log_probs')

  self.log_probs = attributes[:'log_probs'] if attributes[:'log_probs']

  self.max_tokens = attributes[:'maxTokens'] if attributes[:'maxTokens']

  raise 'You cannot provide both :maxTokens and :max_tokens' if attributes.key?(:'maxTokens') && attributes.key?(:'max_tokens')

  self.max_tokens = attributes[:'max_tokens'] if attributes[:'max_tokens']
end

Instance Attribute Details

#frequency_penalty ⇒ Float

To reduce repetitiveness of generated tokens, this number penalizes new tokens based on their frequency in the generated text so far. Values > 0 encourage the model to use new tokens and values < 0 encourage the model to repeat tokens. Set to 0 to disable.

Returns:

  • (Float) 


51
52
53
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 51

def frequency_penalty
  @frequency_penalty
end

#is_echo ⇒ BOOLEAN

Whether or not to return the user prompt in the response. Applies only to non-stream results.

Returns:

  • (BOOLEAN) 


26
27
28
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 26

def is_echo
  @is_echo
end

#is_stream ⇒ BOOLEAN

Whether to stream back partial progress. If set, tokens are sent as data-only server-sent events as they become available.

Returns:

  • (BOOLEAN) 


18
19
20
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 18

def is_stream
  @is_stream
end

#log_probs ⇒ Integer

Includes the logarithmic probabilities for the most likely output tokens and the chosen tokens.

For example, if the log probability is 5, the API returns a list of the 5 most likely tokens. The API returns the log probability of the sampled token, so there might be up to logprobs+1 elements in the response.

Returns:

  • (Integer) 


69
70
71
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 69

def log_probs
  @log_probs
end

#max_tokens ⇒ Integer

The maximum number of tokens that can be generated per output sequence. The token count of your prompt plus max_tokens cannot exceed the model's context length.

Returns:

  • (Integer) 


73
74
75
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 73

def max_tokens
  @max_tokens
end

#num_generations ⇒ Integer

The number of of generated texts that will be returned.

Returns:

  • (Integer) 


22
23
24
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 22

def num_generations
  @num_generations
end

#presence_penalty ⇒ Float

To reduce repetitiveness of generated tokens, this number penalizes new tokens based on whether they've appeared in the generated text so far. Values > 0 encourage the model to use new tokens and values < 0 encourage the model to repeat tokens.

Similar to frequency penalty, a penalty is applied to previously present tokens, except that this penalty is applied equally to all tokens that have already appeared, regardless of how many times they've appeared. Set to 0 to disable.

Returns:

  • (Float) 


58
59
60
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 58

def presence_penalty
  @presence_penalty
end

#prompt ⇒ String

Represents the prompt to be completed. The trailing white spaces are trimmed before completion.

Returns:

  • (String) 


14
15
16
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 14

def prompt
  @prompt
end

#stop ⇒ Array<String>

List of strings that stop the generation if they are generated for the response text. The returned output will not contain the stop strings.

Returns:

  • (Array<String>) 


62
63
64
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 62

def stop
  @stop
end

#temperature ⇒ Float

A number that sets the randomness of the generated output. A lower temperature means a less random generations.

Use lower numbers for tasks with a correct answer such as question answering or summarizing. High temperatures can generate hallucinations or factually incorrect information. Start with temperatures lower than 1.0 and increase the temperature for more creative outputs, as you regenerate the prompts to refine the outputs.

Returns:

  • (Float) 


47
48
49
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 47

def temperature
  @temperature
end

#top_k ⇒ Integer

An integer that sets up the model to use only the top k most likely tokens in the generated output. A higher k introduces more randomness into the output making the output text sound more natural. Default value is -1 which means to consider all tokens. Setting to 0 disables this method and considers all tokens.

If also using top p, then the model considers only the top tokens whose probabilities add up to p percent and ignores the rest of the k tokens. For example, if k is 20, but the probabilities of the top 10 add up to .75, then only the top 10 tokens are chosen.

Returns:

  • (Integer) 


33
34
35
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 33

def top_k
  @top_k
end

#top_p ⇒ Float

If set to a probability 0.0 < p < 1.0, it ensures that only the most likely tokens, with total probability mass of p, are considered for generation at each step.

To eliminate tokens with low likelihood, assign p a minimum percentage for the next token's likelihood. For example, when p is set to 0.75, the model eliminates the bottom 25 percent for the next token. Set to 1 to consider all tokens and set to 0 to disable. If both k and p are enabled, p acts after k.

Returns:

  • (Float) 


40
41
42
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 40

def top_p
  @top_p
end

Class Method Details

.attribute_map ⇒ Object

Attribute mapping from ruby-style variable name to JSON key.



76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 76

def self.attribute_map
  {
    # rubocop:disable Style/SymbolLiteral
    'runtime_type': :'runtimeType',
    'prompt': :'prompt',
    'is_stream': :'isStream',
    'num_generations': :'numGenerations',
    'is_echo': :'isEcho',
    'top_k': :'topK',
    'top_p': :'topP',
    'temperature': :'temperature',
    'frequency_penalty': :'frequencyPenalty',
    'presence_penalty': :'presencePenalty',
    'stop': :'stop',
    'log_probs': :'logProbs',
    'max_tokens': :'maxTokens'
    # rubocop:enable Style/SymbolLiteral
  }
end

.swagger_types ⇒ Object

Attribute type mapping.



97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 97

def self.swagger_types
  {
    # rubocop:disable Style/SymbolLiteral
    'runtime_type': :'String',
    'prompt': :'String',
    'is_stream': :'BOOLEAN',
    'num_generations': :'Integer',
    'is_echo': :'BOOLEAN',
    'top_k': :'Integer',
    'top_p': :'Float',
    'temperature': :'Float',
    'frequency_penalty': :'Float',
    'presence_penalty': :'Float',
    'stop': :'Array<String>',
    'log_probs': :'Integer',
    'max_tokens': :'Integer'
    # rubocop:enable Style/SymbolLiteral
  }
end

Instance Method Details

#==(other) ⇒ Object

Checks equality by comparing each attribute.

Parameters:

  • other (Object) —

    the other object to be compared



224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 224

def ==(other)
  return true if equal?(other)

  self.class == other.class &&
    runtime_type == other.runtime_type &&
    prompt == other.prompt &&
    is_stream == other.is_stream &&
    num_generations == other.num_generations &&
    is_echo == other.is_echo &&
    top_k == other.top_k &&
    top_p == other.top_p &&
    temperature == other.temperature &&
    frequency_penalty == other.frequency_penalty &&
    presence_penalty == other.presence_penalty &&
    stop == other.stop &&
    log_probs == other.log_probs &&
    max_tokens == other.max_tokens
end

#build_from_hash(attributes) ⇒ Object

Builds the object from hash

Parameters:

  • attributes (Hash) —

    Model attributes in the form of hash

Returns:

  • (Object) —

    Returns the model itself



266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 266

def build_from_hash(attributes)
  return nil unless attributes.is_a?(Hash)

  self.class.swagger_types.each_pair do |key, type|
    if type =~ /^Array<(.*)>/i
      # check to ensure the input is an array given that the the attribute
      # is documented as an array but the input is not
      if attributes[self.class.attribute_map[key]].is_a?(Array)
        public_method("#{key}=").call(
          attributes[self.class.attribute_map[key]]
            .map { |v| OCI::Internal::Util.convert_to_type(Regexp.last_match(1), v) }
        )
      end
    elsif !attributes[self.class.attribute_map[key]].nil?
      public_method("#{key}=").call(
        OCI::Internal::Util.convert_to_type(type, attributes[self.class.attribute_map[key]])
      )
    end
    # or else data not found in attributes(hash), not an issue as the data can be optional
  end

  self
end

#eql?(other) ⇒ Boolean

Parameters:

  • other (Object) —

    the other object to be compared

Returns:

  • (Boolean)

See Also:

  • `==` method


246
247
248
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 246

def eql?(other)
  self == other
end

#hash ⇒ Fixnum

Calculates hash code according to all attributes.

Returns:

  • (Fixnum) —

    Hash code



255
256
257
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 255

def hash
  [runtime_type, prompt, is_stream, num_generations, is_echo, top_k, top_p, temperature, frequency_penalty, presence_penalty, stop, log_probs, max_tokens].hash
end

#to_hash ⇒ Hash

Returns the object in the form of hash

Returns:

  • (Hash) —

    Returns the object in the form of hash



299
300
301
302
303
304
305
306
307
308
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 299

def to_hash
  hash = {}
  self.class.attribute_map.each_pair do |attr, param|
    value = public_method(attr).call
    next if value.nil? && !instance_variable_defined?("@#{attr}")

    hash[param] = _to_hash(value)
  end
  hash
end

#to_s ⇒ String

Returns the string representation of the object

Returns:

  • (String) —

    String presentation of the object



293
294
295
 
# File 'lib/oci/generative_ai_inference/models/llama_llm_inference_request.rb', line 293

def to_s
  to_hash.to_s
end