Class: Google::Apis::AiplatformV1::GoogleCloudAiplatformV1SchemaModelevaluationMetricsClassificationEvaluationMetricsConfidenceMetrics

Inherits:
Object
  • Object
show all
Includes:
Core::Hashable, Core::JsonObjectSupport
Defined in:
lib/google/apis/aiplatform_v1/classes.rb,
lib/google/apis/aiplatform_v1/representations.rb,
lib/google/apis/aiplatform_v1/representations.rb

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(**args) ⇒ GoogleCloudAiplatformV1SchemaModelevaluationMetricsClassificationEvaluationMetricsConfidenceMetrics

Returns a new instance of GoogleCloudAiplatformV1SchemaModelevaluationMetricsClassificationEvaluationMetricsConfidenceMetrics.



20171
20172
20173
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20171

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

Instance Attribute Details

#confidence_threshold ⇒ Float

Metrics are computed with an assumption that the Model never returns predictions with score lower than this value. Corresponds to the JSON property confidenceThreshold

Returns:

  • (Float) 


20080
20081
20082
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20080

def confidence_threshold
  @confidence_threshold
end

#confusion_matrix ⇒ Google::Apis::AiplatformV1::GoogleCloudAiplatformV1SchemaModelevaluationMetricsConfusionMatrix

Confusion matrix of the evaluation for this confidence_threshold. Corresponds to the JSON property confusionMatrix

Returns:



20085
20086
20087
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20085

def confusion_matrix
  @confusion_matrix
end

#f1_score ⇒ Float

The harmonic mean of recall and precision. For summary metrics, it computes the micro-averaged F1 score. Corresponds to the JSON property f1Score

Returns:

  • (Float) 


20091
20092
20093
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20091

def f1_score
  @f1_score
end

#f1_score_at1 ⇒ Float

The harmonic mean of recallAt1 and precisionAt1. Corresponds to the JSON property f1ScoreAt1

Returns:

  • (Float) 


20096
20097
20098
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20096

def f1_score_at1
  @f1_score_at1
end

#f1_score_macro ⇒ Float

Macro-averaged F1 Score. Corresponds to the JSON property f1ScoreMacro

Returns:

  • (Float) 


20101
20102
20103
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20101

def f1_score_macro
  @f1_score_macro
end

#f1_score_micro ⇒ Float

Micro-averaged F1 Score. Corresponds to the JSON property f1ScoreMicro

Returns:

  • (Float) 


20106
20107
20108
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20106

def f1_score_micro
  @f1_score_micro
end

#false_negative_count ⇒ Fixnum

The number of ground truth labels that are not matched by a Model created label. Corresponds to the JSON property falseNegativeCount

Returns:

  • (Fixnum) 


20112
20113
20114
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20112

def false_negative_count
  @false_negative_count
end

#false_positive_count ⇒ Fixnum

The number of Model created labels that do not match a ground truth label. Corresponds to the JSON property falsePositiveCount

Returns:

  • (Fixnum) 


20117
20118
20119
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20117

def false_positive_count
  @false_positive_count
end

#false_positive_rate ⇒ Float

False Positive Rate for the given confidence threshold. Corresponds to the JSON property falsePositiveRate

Returns:

  • (Float) 


20122
20123
20124
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20122

def false_positive_rate
  @false_positive_rate
end

#false_positive_rate_at1 ⇒ Float

The False Positive Rate when only considering the label that has the highest prediction score and not below the confidence threshold for each DataItem. Corresponds to the JSON property falsePositiveRateAt1

Returns:

  • (Float) 


20128
20129
20130
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20128

def false_positive_rate_at1
  @false_positive_rate_at1
end

#max_predictions ⇒ Fixnum

Metrics are computed with an assumption that the Model always returns at most this many predictions (ordered by their score, descendingly), but they all still need to meet the confidenceThreshold. Corresponds to the JSON property maxPredictions

Returns:

  • (Fixnum) 


20135
20136
20137
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20135

def max_predictions
  @max_predictions
end

#precision ⇒ Float

Precision for the given confidence threshold. Corresponds to the JSON property precision

Returns:

  • (Float) 


20140
20141
20142
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20140

def precision
  @precision
end

#precision_at1 ⇒ Float

The precision when only considering the label that has the highest prediction score and not below the confidence threshold for each DataItem. Corresponds to the JSON property precisionAt1

Returns:

  • (Float) 


20146
20147
20148
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20146

def precision_at1
  @precision_at1
end

#recall ⇒ Float

Recall (True Positive Rate) for the given confidence threshold. Corresponds to the JSON property recall

Returns:

  • (Float) 


20151
20152
20153
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20151

def recall
  @recall
end

#recall_at1 ⇒ Float

The Recall (True Positive Rate) when only considering the label that has the highest prediction score and not below the confidence threshold for each DataItem. Corresponds to the JSON property recallAt1

Returns:

  • (Float) 


20158
20159
20160
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20158

def recall_at1
  @recall_at1
end

#true_negative_count ⇒ Fixnum

The number of labels that were not created by the Model, but if they would, they would not match a ground truth label. Corresponds to the JSON property trueNegativeCount

Returns:

  • (Fixnum) 


20164
20165
20166
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20164

def true_negative_count
  @true_negative_count
end

#true_positive_count ⇒ Fixnum

The number of Model created labels that match a ground truth label. Corresponds to the JSON property truePositiveCount

Returns:

  • (Fixnum) 


20169
20170
20171
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20169

def true_positive_count
  @true_positive_count
end

Instance Method Details

#update!(**args) ⇒ Object

Update properties of this object



20176
20177
20178
20179
20180
20181
20182
20183
20184
20185
20186
20187
20188
20189
20190
20191
20192
20193
20194
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 20176

def update!(**args)
  @confidence_threshold = args[:confidence_threshold] if args.key?(:confidence_threshold)
  @confusion_matrix = args[:confusion_matrix] if args.key?(:confusion_matrix)
  @f1_score = args[:f1_score] if args.key?(:f1_score)
  @f1_score_at1 = args[:f1_score_at1] if args.key?(:f1_score_at1)
  @f1_score_macro = args[:f1_score_macro] if args.key?(:f1_score_macro)
  @f1_score_micro = args[:f1_score_micro] if args.key?(:f1_score_micro)
  @false_negative_count = args[:false_negative_count] if args.key?(:false_negative_count)
  @false_positive_count = args[:false_positive_count] if args.key?(:false_positive_count)
  @false_positive_rate = args[:false_positive_rate] if args.key?(:false_positive_rate)
  @false_positive_rate_at1 = args[:false_positive_rate_at1] if args.key?(:false_positive_rate_at1)
  @max_predictions = args[:max_predictions] if args.key?(:max_predictions)
  @precision = args[:precision] if args.key?(:precision)
  @precision_at1 = args[:precision_at1] if args.key?(:precision_at1)
  @recall = args[:recall] if args.key?(:recall)
  @recall_at1 = args[:recall_at1] if args.key?(:recall_at1)
  @true_negative_count = args[:true_negative_count] if args.key?(:true_negative_count)
  @true_positive_count = args[:true_positive_count] if args.key?(:true_positive_count)
end