Class: Google::Cloud::AIPlatform::V1::EvaluateInstancesResponse

Inherits:

Object

• Object
• Google::Cloud::AIPlatform::V1::EvaluateInstancesResponse

Extended by:

Protobuf::MessageExts::ClassMethods

Includes:

Protobuf::MessageExts

Defined in:

proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb

Overview

Response message for EvaluationService.EvaluateInstances.

Instance Attribute Summary

• #bleu_results ⇒ ::Google::Cloud::AIPlatform::V1::BleuResults

  Results for bleu metric.

• #coherence_result ⇒ ::Google::Cloud::AIPlatform::V1::CoherenceResult

  Result for coherence metric.

• #comet_result ⇒ ::Google::Cloud::AIPlatform::V1::CometResult

  Translation metrics.

• #exact_match_results ⇒ ::Google::Cloud::AIPlatform::V1::ExactMatchResults

  Auto metric evaluation results.

• #fluency_result ⇒ ::Google::Cloud::AIPlatform::V1::FluencyResult

  LLM-based metric evaluation result.

• #fulfillment_result ⇒ ::Google::Cloud::AIPlatform::V1::FulfillmentResult

  Result for fulfillment metric.

• #groundedness_result ⇒ ::Google::Cloud::AIPlatform::V1::GroundednessResult

  Result for groundedness metric.

• #metric_results ⇒ ::Array<::Google::Cloud::AIPlatform::V1::MetricResult>

  Metric results for each instance.

• #metricx_result ⇒ ::Google::Cloud::AIPlatform::V1::MetricxResult

  Result for Metricx metric.

• #pairwise_metric_result ⇒ ::Google::Cloud::AIPlatform::V1::PairwiseMetricResult

  Result for pairwise metric.

• #pairwise_question_answering_quality_result ⇒ ::Google::Cloud::AIPlatform::V1::PairwiseQuestionAnsweringQualityResult

  Result for pairwise question answering quality metric.

• #pairwise_summarization_quality_result ⇒ ::Google::Cloud::AIPlatform::V1::PairwiseSummarizationQualityResult

  Result for pairwise summarization quality metric.

• #pointwise_metric_result ⇒ ::Google::Cloud::AIPlatform::V1::PointwiseMetricResult

  Generic metrics.

• #question_answering_correctness_result ⇒ ::Google::Cloud::AIPlatform::V1::QuestionAnsweringCorrectnessResult

  Result for question answering correctness metric.

• #question_answering_helpfulness_result ⇒ ::Google::Cloud::AIPlatform::V1::QuestionAnsweringHelpfulnessResult

  Result for question answering helpfulness metric.

• #question_answering_quality_result ⇒ ::Google::Cloud::AIPlatform::V1::QuestionAnsweringQualityResult

  Question answering only metrics.

• #question_answering_relevance_result ⇒ ::Google::Cloud::AIPlatform::V1::QuestionAnsweringRelevanceResult

  Result for question answering relevance metric.

• #rouge_results ⇒ ::Google::Cloud::AIPlatform::V1::RougeResults

  Results for rouge metric.

• #safety_result ⇒ ::Google::Cloud::AIPlatform::V1::SafetyResult

  Result for safety metric.

• #summarization_helpfulness_result ⇒ ::Google::Cloud::AIPlatform::V1::SummarizationHelpfulnessResult

  Result for summarization helpfulness metric.

• #summarization_quality_result ⇒ ::Google::Cloud::AIPlatform::V1::SummarizationQualityResult

  Summarization only metrics.

• #summarization_verbosity_result ⇒ ::Google::Cloud::AIPlatform::V1::SummarizationVerbosityResult

  Result for summarization verbosity metric.

• #tool_call_valid_results ⇒ ::Google::Cloud::AIPlatform::V1::ToolCallValidResults

  Tool call metrics.

• #tool_name_match_results ⇒ ::Google::Cloud::AIPlatform::V1::ToolNameMatchResults

  Results for tool name match metric.

• #tool_parameter_key_match_results ⇒ ::Google::Cloud::AIPlatform::V1::ToolParameterKeyMatchResults

  Results for tool parameter key match metric.

• #tool_parameter_kv_match_results ⇒ ::Google::Cloud::AIPlatform::V1::ToolParameterKVMatchResults

  Results for tool parameter key value match metric.

Instance Attribute Details

#bleu_results ⇒ ::Google::Cloud::AIPlatform::V1::BleuResults

Returns Results for bleu metric.

Note: The following fields are mutually exclusive: bleu_results, exact_match_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::BleuResults) —

  Results for bleu metric.

  Note: The following fields are mutually exclusive: bleu_results, exact_match_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#coherence_result ⇒ ::Google::Cloud::AIPlatform::V1::CoherenceResult

Returns Result for coherence metric.

Note: The following fields are mutually exclusive: coherence_result, exact_match_results, bleu_results, rouge_results, fluency_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::CoherenceResult) —

  Result for coherence metric.

  Note: The following fields are mutually exclusive: coherence_result, exact_match_results, bleu_results, rouge_results, fluency_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#comet_result ⇒ ::Google::Cloud::AIPlatform::V1::CometResult

Returns Translation metrics. Result for Comet metric.

Note: The following fields are mutually exclusive: comet_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::CometResult) —

  Translation metrics. Result for Comet metric.

  Note: The following fields are mutually exclusive: comet_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#exact_match_results ⇒ ::Google::Cloud::AIPlatform::V1::ExactMatchResults

Returns Auto metric evaluation results. Results for exact match metric.

Note: The following fields are mutually exclusive: exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::ExactMatchResults) —

  Auto metric evaluation results. Results for exact match metric.

  Note: The following fields are mutually exclusive: exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#fluency_result ⇒ ::Google::Cloud::AIPlatform::V1::FluencyResult

Returns LLM-based metric evaluation result. General text generation metrics, applicable to other categories. Result for fluency metric.

Note: The following fields are mutually exclusive: fluency_result, exact_match_results, bleu_results, rouge_results, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::FluencyResult) —

  LLM-based metric evaluation result. General text generation metrics, applicable to other categories. Result for fluency metric.

  Note: The following fields are mutually exclusive: fluency_result, exact_match_results, bleu_results, rouge_results, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#fulfillment_result ⇒ ::Google::Cloud::AIPlatform::V1::FulfillmentResult

Returns Result for fulfillment metric.

Note: The following fields are mutually exclusive: fulfillment_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::FulfillmentResult) —

  Result for fulfillment metric.

  Note: The following fields are mutually exclusive: fulfillment_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#groundedness_result ⇒ ::Google::Cloud::AIPlatform::V1::GroundednessResult

Returns Result for groundedness metric.

Note: The following fields are mutually exclusive: groundedness_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::GroundednessResult) —

  Result for groundedness metric.

  Note: The following fields are mutually exclusive: groundedness_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#metric_results ⇒ ::Array<::Google::Cloud::AIPlatform::V1::MetricResult>

Returns Metric results for each instance. The order of the metric results is guaranteed to be the same as the order of the instances in the request.

Returns:

• (::Array<::Google::Cloud::AIPlatform::V1::MetricResult>) —

  Metric results for each instance. The order of the metric results is guaranteed to be the same as the order of the instances in the request.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#metricx_result ⇒ ::Google::Cloud::AIPlatform::V1::MetricxResult

Returns Result for Metricx metric.

Note: The following fields are mutually exclusive: metricx_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::MetricxResult) —

  Result for Metricx metric.

  Note: The following fields are mutually exclusive: metricx_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#pairwise_metric_result ⇒ ::Google::Cloud::AIPlatform::V1::PairwiseMetricResult

Returns Result for pairwise metric.

Note: The following fields are mutually exclusive: pairwise_metric_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::PairwiseMetricResult) —

  Result for pairwise metric.

  Note: The following fields are mutually exclusive: pairwise_metric_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#pairwise_question_answering_quality_result ⇒ ::Google::Cloud::AIPlatform::V1::PairwiseQuestionAnsweringQualityResult

Returns Result for pairwise question answering quality metric.

Note: The following fields are mutually exclusive: pairwise_question_answering_quality_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::PairwiseQuestionAnsweringQualityResult) —

  Result for pairwise question answering quality metric.

  Note: The following fields are mutually exclusive: pairwise_question_answering_quality_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#pairwise_summarization_quality_result ⇒ ::Google::Cloud::AIPlatform::V1::PairwiseSummarizationQualityResult

Returns Result for pairwise summarization quality metric.

Note: The following fields are mutually exclusive: pairwise_summarization_quality_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::PairwiseSummarizationQualityResult) —

  Result for pairwise summarization quality metric.

  Note: The following fields are mutually exclusive: pairwise_summarization_quality_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#pointwise_metric_result ⇒ ::Google::Cloud::AIPlatform::V1::PointwiseMetricResult

Returns Generic metrics. Result for pointwise metric.

Note: The following fields are mutually exclusive: pointwise_metric_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::PointwiseMetricResult) —

  Generic metrics. Result for pointwise metric.

  Note: The following fields are mutually exclusive: pointwise_metric_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#question_answering_correctness_result ⇒ ::Google::Cloud::AIPlatform::V1::QuestionAnsweringCorrectnessResult

Returns Result for question answering correctness metric.

Note: The following fields are mutually exclusive: question_answering_correctness_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::QuestionAnsweringCorrectnessResult) —

  Result for question answering correctness metric.

  Note: The following fields are mutually exclusive: question_answering_correctness_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#question_answering_helpfulness_result ⇒ ::Google::Cloud::AIPlatform::V1::QuestionAnsweringHelpfulnessResult

Returns Result for question answering helpfulness metric.

Note: The following fields are mutually exclusive: question_answering_helpfulness_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::QuestionAnsweringHelpfulnessResult) —

  Result for question answering helpfulness metric.

  Note: The following fields are mutually exclusive: question_answering_helpfulness_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#question_answering_quality_result ⇒ ::Google::Cloud::AIPlatform::V1::QuestionAnsweringQualityResult

Returns Question answering only metrics. Result for question answering quality metric.

Note: The following fields are mutually exclusive: question_answering_quality_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::QuestionAnsweringQualityResult) —

  Question answering only metrics. Result for question answering quality metric.

  Note: The following fields are mutually exclusive: question_answering_quality_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#question_answering_relevance_result ⇒ ::Google::Cloud::AIPlatform::V1::QuestionAnsweringRelevanceResult

Returns Result for question answering relevance metric.

Note: The following fields are mutually exclusive: question_answering_relevance_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::QuestionAnsweringRelevanceResult) —

  Result for question answering relevance metric.

  Note: The following fields are mutually exclusive: question_answering_relevance_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#rouge_results ⇒ ::Google::Cloud::AIPlatform::V1::RougeResults

Returns Results for rouge metric.

Note: The following fields are mutually exclusive: rouge_results, exact_match_results, bleu_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::RougeResults) —

  Results for rouge metric.

  Note: The following fields are mutually exclusive: rouge_results, exact_match_results, bleu_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#safety_result ⇒ ::Google::Cloud::AIPlatform::V1::SafetyResult

Returns Result for safety metric.

Note: The following fields are mutually exclusive: safety_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::SafetyResult) —

  Result for safety metric.

  Note: The following fields are mutually exclusive: safety_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#summarization_helpfulness_result ⇒ ::Google::Cloud::AIPlatform::V1::SummarizationHelpfulnessResult

Returns Result for summarization helpfulness metric.

Note: The following fields are mutually exclusive: summarization_helpfulness_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::SummarizationHelpfulnessResult) —

  Result for summarization helpfulness metric.

  Note: The following fields are mutually exclusive: summarization_helpfulness_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#summarization_quality_result ⇒ ::Google::Cloud::AIPlatform::V1::SummarizationQualityResult

Returns Summarization only metrics. Result for summarization quality metric.

Note: The following fields are mutually exclusive: summarization_quality_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::SummarizationQualityResult) —

  Summarization only metrics. Result for summarization quality metric.

  Note: The following fields are mutually exclusive: summarization_quality_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#summarization_verbosity_result ⇒ ::Google::Cloud::AIPlatform::V1::SummarizationVerbosityResult

Returns Result for summarization verbosity metric.

Note: The following fields are mutually exclusive: summarization_verbosity_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::SummarizationVerbosityResult) —

  Result for summarization verbosity metric.

  Note: The following fields are mutually exclusive: summarization_verbosity_result, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#tool_call_valid_results ⇒ ::Google::Cloud::AIPlatform::V1::ToolCallValidResults

Returns Tool call metrics. Results for tool call valid metric.

Note: The following fields are mutually exclusive: tool_call_valid_results, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::ToolCallValidResults) —

  Tool call metrics. Results for tool call valid metric.

  Note: The following fields are mutually exclusive: tool_call_valid_results, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_name_match_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#tool_name_match_results ⇒ ::Google::Cloud::AIPlatform::V1::ToolNameMatchResults

Returns Results for tool name match metric.

Note: The following fields are mutually exclusive: tool_name_match_results, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::ToolNameMatchResults) —

  Results for tool name match metric.

  Note: The following fields are mutually exclusive: tool_name_match_results, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_parameter_key_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#tool_parameter_key_match_results ⇒ ::Google::Cloud::AIPlatform::V1::ToolParameterKeyMatchResults

Returns Results for tool parameter key match metric.

Note: The following fields are mutually exclusive: tool_parameter_key_match_results, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::ToolParameterKeyMatchResults) —

  Results for tool parameter key match metric.

  Note: The following fields are mutually exclusive: tool_parameter_key_match_results, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_kv_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end

#tool_parameter_kv_match_results ⇒ ::Google::Cloud::AIPlatform::V1::ToolParameterKVMatchResults

Returns Results for tool parameter key value match metric.

Note: The following fields are mutually exclusive: tool_parameter_kv_match_results, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

Returns:

• (::Google::Cloud::AIPlatform::V1::ToolParameterKVMatchResults) —

  Results for tool parameter key value match metric.

  Note: The following fields are mutually exclusive: tool_parameter_kv_match_results, exact_match_results, bleu_results, rouge_results, fluency_result, coherence_result, safety_result, groundedness_result, fulfillment_result, summarization_quality_result, pairwise_summarization_quality_result, summarization_helpfulness_result, summarization_verbosity_result, question_answering_quality_result, pairwise_question_answering_quality_result, question_answering_relevance_result, question_answering_helpfulness_result, question_answering_correctness_result, pointwise_metric_result, pairwise_metric_result, tool_call_valid_results, tool_name_match_results, tool_parameter_key_match_results, comet_result, metricx_result. If a field in that set is populated, all other fields in the set will automatically be cleared.

# File 'proto_docs/google/cloud/aiplatform/v1/evaluation_service.rb', line 424

class EvaluateInstancesResponse
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods
end