Class: Google::Cloud::DataLabeling::V1beta1::PrCurve::ConfidenceMetricsEntry

Inherits:

Object

• Object
• Google::Cloud::DataLabeling::V1beta1::PrCurve::ConfidenceMetricsEntry

Extended by:

Protobuf::MessageExts::ClassMethods

Includes:

Protobuf::MessageExts

Defined in:

proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb

Instance Attribute Summary

• #confidence_threshold ⇒ ::Float

  Threshold used for this entry.

• #f1_score ⇒ ::Float

  Harmonic mean of recall and precision.

• #f1_score_at1 ⇒ ::Float

  The harmonic mean of recall_at1 and precision_at1.

• #f1_score_at5 ⇒ ::Float

  The harmonic mean of recall_at5 and precision_at5.

• #precision ⇒ ::Float

  Precision value.

• #precision_at1 ⇒ ::Float

  Precision value for entries with label that has highest score.

• #precision_at5 ⇒ ::Float

  Precision value for entries with label that has highest 5 scores.

• #recall ⇒ ::Float

  Recall value.

• #recall_at1 ⇒ ::Float

  Recall value for entries with label that has highest score.

• #recall_at5 ⇒ ::Float

  Recall value for entries with label that has highest 5 scores.

Instance Attribute Details

#confidence_threshold ⇒ ::Float

Returns Threshold used for this entry.

For classification tasks, this is a classification threshold: a predicted label is categorized as positive or negative (in the context of this point on the PR curve) based on whether the label's score meets this threshold.

For image object detection (bounding box) tasks, this is the [intersection-over-union

(IOU)](/vision/automl/object-detection/docs/evaluate#intersection-over-union) threshold for the context of this point on the PR curve.

Returns:

• (::Float) —

  Threshold used for this entry.

  For classification tasks, this is a classification threshold: a predicted label is categorized as positive or negative (in the context of this point on the PR curve) based on whether the label's score meets this threshold.

  For image object detection (bounding box) tasks, this is the [intersection-over-union

  (IOU)](/vision/automl/object-detection/docs/evaluate#intersection-over-union) threshold for the context of this point on the PR curve.

# File 'proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb', line 181

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

#f1_score ⇒ ::Float

Returns Harmonic mean of recall and precision.

Returns:

• (::Float) —

  Harmonic mean of recall and precision.

# File 'proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb', line 181

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

#f1_score_at1 ⇒ ::Float

Returns The harmonic mean of recall_at1 and precision_at1.

Returns:

• (::Float) —

  The harmonic mean of recall_at1 and precision_at1.

# File 'proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb', line 181

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

#f1_score_at5 ⇒ ::Float

Returns The harmonic mean of recall_at5 and precision_at5.

Returns:

• (::Float) —

  The harmonic mean of recall_at5 and precision_at5.

# File 'proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb', line 181

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

#precision ⇒ ::Float

Returns Precision value.

Returns:

• (::Float) —

  Precision value.

# File 'proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb', line 181

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

#precision_at1 ⇒ ::Float

Returns Precision value for entries with label that has highest score.

Returns:

• (::Float) —

  Precision value for entries with label that has highest score.

# File 'proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb', line 181

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

#precision_at5 ⇒ ::Float

Returns Precision value for entries with label that has highest 5 scores.

Returns:

• (::Float) —

  Precision value for entries with label that has highest 5 scores.

# File 'proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb', line 181

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

#recall ⇒ ::Float

Returns Recall value.

Returns:

• (::Float) —

  Recall value.

# File 'proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb', line 181

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

#recall_at1 ⇒ ::Float

Returns Recall value for entries with label that has highest score.

Returns:

• (::Float) —

  Recall value for entries with label that has highest score.

# File 'proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb', line 181

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

#recall_at5 ⇒ ::Float

Returns Recall value for entries with label that has highest 5 scores.

Returns:

• (::Float) —

  Recall value for entries with label that has highest 5 scores.

# File 'proto_docs/google/cloud/datalabeling/v1beta1/evaluation.rb', line 181

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