Class: Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig

Inherits:
Object
  • Object
show all
Extended by:
Protobuf::MessageExts::ClassMethods
Includes:
Protobuf::MessageExts
Defined in:
proto_docs/google/cloud/aiplatform/v1/model_monitoring.rb

Overview

The objective configuration for model monitoring, including the information needed to detect anomalies for one particular model.

Defined Under Namespace

Classes: ExplanationConfig, PredictionDriftDetectionConfig, TrainingDataset, TrainingPredictionSkewDetectionConfig

Instance Attribute Summary collapse

Instance Attribute Details

#explanation_config ⇒ ::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::ExplanationConfig

Returns The config for integrating with Vertex Explainable AI.

Returns:



39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
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
 
# File 'proto_docs/google/cloud/aiplatform/v1/model_monitoring.rb', line 39

class ModelMonitoringObjectiveConfig
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods

  # Training Dataset information.
  # @!attribute [rw] dataset
  #   @return [::String]
  #     The resource name of the Dataset used to train this Model.
  # @!attribute [rw] gcs_source
  #   @return [::Google::Cloud::AIPlatform::V1::GcsSource]
  #     The Google Cloud Storage uri of the unmanaged Dataset used to train
  #     this Model.
  # @!attribute [rw] bigquery_source
  #   @return [::Google::Cloud::AIPlatform::V1::BigQuerySource]
  #     The BigQuery table of the unmanaged Dataset used to train this
  #     Model.
  # @!attribute [rw] data_format
  #   @return [::String]
  #     Data format of the dataset, only applicable if the input is from
  #     Google Cloud Storage.
  #     The possible formats are:
  #
  #     "tf-record"
  #     The source file is a TFRecord file.
  #
  #     "csv"
  #     The source file is a CSV file.
  #     "jsonl"
  #     The source file is a JSONL file.
  # @!attribute [rw] target_field
  #   @return [::String]
  #     The target field name the model is to predict.
  #     This field will be excluded when doing Predict and (or) Explain for the
  #     training data.
  # @!attribute [rw] logging_sampling_strategy
  #   @return [::Google::Cloud::AIPlatform::V1::SamplingStrategy]
  #     Strategy to sample data from Training Dataset.
  #     If not set, we process the whole dataset.
  class TrainingDataset
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods
  end

  # The config for Training & Prediction data skew detection. It specifies the
  # training dataset sources and the skew detection parameters.
  # @!attribute [rw] skew_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. If a feature needs to
  #     be monitored for skew, a value threshold must be configured for that
  #     feature. The threshold here is against feature distribution distance
  #     between the training and prediction feature.
  # @!attribute [rw] attribution_score_skew_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. The threshold here is
  #     against attribution score distance between the training and prediction
  #     feature.
  # @!attribute [rw] default_skew_threshold
  #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
  #     Skew anomaly detection threshold used by all features.
  #     When the per-feature thresholds are not set, this field can be used to
  #     specify a threshold for all features.
  class TrainingPredictionSkewDetectionConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class SkewThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class AttributionScoreSkewThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end
  end

  # The config for Prediction data drift detection.
  # @!attribute [rw] drift_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. If a feature needs to
  #     be monitored for drift, a value threshold must be configured for that
  #     feature. The threshold here is against feature distribution distance
  #     between different time windws.
  # @!attribute [rw] attribution_score_drift_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. The threshold here is
  #     against attribution score distance between different time windows.
  # @!attribute [rw] default_drift_threshold
  #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
  #     Drift anomaly detection threshold used by all features.
  #     When the per-feature thresholds are not set, this field can be used to
  #     specify a threshold for all features.
  class PredictionDriftDetectionConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class DriftThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class AttributionScoreDriftThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end
  end

  # The config for integrating with Vertex Explainable AI. Only applicable if
  # the Model has explanation_spec populated.
  # @!attribute [rw] enable_feature_attributes
  #   @return [::Boolean]
  #     If want to analyze the Vertex Explainable AI feature attribute scores or
  #     not. If set to true, Vertex AI will log the feature attributions from
  #     explain response and do the skew/drift detection for them.
  # @!attribute [rw] explanation_baseline
  #   @return [::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::ExplanationConfig::ExplanationBaseline]
  #     Predictions generated by the BatchPredictionJob using baseline dataset.
  class ExplanationConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # Output from
    # {::Google::Cloud::AIPlatform::V1::BatchPredictionJob BatchPredictionJob} for
    # Model Monitoring baseline dataset, which can be used to generate baseline
    # attribution scores.
    # @!attribute [rw] gcs
    #   @return [::Google::Cloud::AIPlatform::V1::GcsDestination]
    #     Cloud Storage location for BatchExplain output.
    # @!attribute [rw] bigquery
    #   @return [::Google::Cloud::AIPlatform::V1::BigQueryDestination]
    #     BigQuery location for BatchExplain output.
    # @!attribute [rw] prediction_format
    #   @return [::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::ExplanationConfig::ExplanationBaseline::PredictionFormat]
    #     The storage format of the predictions generated BatchPrediction job.
    class ExplanationBaseline
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods

      # The storage format of the predictions generated BatchPrediction job.
      module PredictionFormat
        # Should not be set.
        PREDICTION_FORMAT_UNSPECIFIED = 0

        # Predictions are in JSONL files.
        JSONL = 2

        # Predictions are in BigQuery.
        BIGQUERY = 3
      end
    end
  end
end

#prediction_drift_detection_config ⇒ ::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::PredictionDriftDetectionConfig

Returns The config for drift of prediction data.

Returns:



39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
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
 
# File 'proto_docs/google/cloud/aiplatform/v1/model_monitoring.rb', line 39

class ModelMonitoringObjectiveConfig
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods

  # Training Dataset information.
  # @!attribute [rw] dataset
  #   @return [::String]
  #     The resource name of the Dataset used to train this Model.
  # @!attribute [rw] gcs_source
  #   @return [::Google::Cloud::AIPlatform::V1::GcsSource]
  #     The Google Cloud Storage uri of the unmanaged Dataset used to train
  #     this Model.
  # @!attribute [rw] bigquery_source
  #   @return [::Google::Cloud::AIPlatform::V1::BigQuerySource]
  #     The BigQuery table of the unmanaged Dataset used to train this
  #     Model.
  # @!attribute [rw] data_format
  #   @return [::String]
  #     Data format of the dataset, only applicable if the input is from
  #     Google Cloud Storage.
  #     The possible formats are:
  #
  #     "tf-record"
  #     The source file is a TFRecord file.
  #
  #     "csv"
  #     The source file is a CSV file.
  #     "jsonl"
  #     The source file is a JSONL file.
  # @!attribute [rw] target_field
  #   @return [::String]
  #     The target field name the model is to predict.
  #     This field will be excluded when doing Predict and (or) Explain for the
  #     training data.
  # @!attribute [rw] logging_sampling_strategy
  #   @return [::Google::Cloud::AIPlatform::V1::SamplingStrategy]
  #     Strategy to sample data from Training Dataset.
  #     If not set, we process the whole dataset.
  class TrainingDataset
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods
  end

  # The config for Training & Prediction data skew detection. It specifies the
  # training dataset sources and the skew detection parameters.
  # @!attribute [rw] skew_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. If a feature needs to
  #     be monitored for skew, a value threshold must be configured for that
  #     feature. The threshold here is against feature distribution distance
  #     between the training and prediction feature.
  # @!attribute [rw] attribution_score_skew_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. The threshold here is
  #     against attribution score distance between the training and prediction
  #     feature.
  # @!attribute [rw] default_skew_threshold
  #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
  #     Skew anomaly detection threshold used by all features.
  #     When the per-feature thresholds are not set, this field can be used to
  #     specify a threshold for all features.
  class TrainingPredictionSkewDetectionConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class SkewThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class AttributionScoreSkewThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end
  end

  # The config for Prediction data drift detection.
  # @!attribute [rw] drift_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. If a feature needs to
  #     be monitored for drift, a value threshold must be configured for that
  #     feature. The threshold here is against feature distribution distance
  #     between different time windws.
  # @!attribute [rw] attribution_score_drift_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. The threshold here is
  #     against attribution score distance between different time windows.
  # @!attribute [rw] default_drift_threshold
  #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
  #     Drift anomaly detection threshold used by all features.
  #     When the per-feature thresholds are not set, this field can be used to
  #     specify a threshold for all features.
  class PredictionDriftDetectionConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class DriftThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class AttributionScoreDriftThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end
  end

  # The config for integrating with Vertex Explainable AI. Only applicable if
  # the Model has explanation_spec populated.
  # @!attribute [rw] enable_feature_attributes
  #   @return [::Boolean]
  #     If want to analyze the Vertex Explainable AI feature attribute scores or
  #     not. If set to true, Vertex AI will log the feature attributions from
  #     explain response and do the skew/drift detection for them.
  # @!attribute [rw] explanation_baseline
  #   @return [::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::ExplanationConfig::ExplanationBaseline]
  #     Predictions generated by the BatchPredictionJob using baseline dataset.
  class ExplanationConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # Output from
    # {::Google::Cloud::AIPlatform::V1::BatchPredictionJob BatchPredictionJob} for
    # Model Monitoring baseline dataset, which can be used to generate baseline
    # attribution scores.
    # @!attribute [rw] gcs
    #   @return [::Google::Cloud::AIPlatform::V1::GcsDestination]
    #     Cloud Storage location for BatchExplain output.
    # @!attribute [rw] bigquery
    #   @return [::Google::Cloud::AIPlatform::V1::BigQueryDestination]
    #     BigQuery location for BatchExplain output.
    # @!attribute [rw] prediction_format
    #   @return [::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::ExplanationConfig::ExplanationBaseline::PredictionFormat]
    #     The storage format of the predictions generated BatchPrediction job.
    class ExplanationBaseline
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods

      # The storage format of the predictions generated BatchPrediction job.
      module PredictionFormat
        # Should not be set.
        PREDICTION_FORMAT_UNSPECIFIED = 0

        # Predictions are in JSONL files.
        JSONL = 2

        # Predictions are in BigQuery.
        BIGQUERY = 3
      end
    end
  end
end

#training_dataset ⇒ ::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::TrainingDataset

Returns Training dataset for models. This field has to be set only if TrainingPredictionSkewDetectionConfig is specified.

Returns:



39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
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
 
# File 'proto_docs/google/cloud/aiplatform/v1/model_monitoring.rb', line 39

class ModelMonitoringObjectiveConfig
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods

  # Training Dataset information.
  # @!attribute [rw] dataset
  #   @return [::String]
  #     The resource name of the Dataset used to train this Model.
  # @!attribute [rw] gcs_source
  #   @return [::Google::Cloud::AIPlatform::V1::GcsSource]
  #     The Google Cloud Storage uri of the unmanaged Dataset used to train
  #     this Model.
  # @!attribute [rw] bigquery_source
  #   @return [::Google::Cloud::AIPlatform::V1::BigQuerySource]
  #     The BigQuery table of the unmanaged Dataset used to train this
  #     Model.
  # @!attribute [rw] data_format
  #   @return [::String]
  #     Data format of the dataset, only applicable if the input is from
  #     Google Cloud Storage.
  #     The possible formats are:
  #
  #     "tf-record"
  #     The source file is a TFRecord file.
  #
  #     "csv"
  #     The source file is a CSV file.
  #     "jsonl"
  #     The source file is a JSONL file.
  # @!attribute [rw] target_field
  #   @return [::String]
  #     The target field name the model is to predict.
  #     This field will be excluded when doing Predict and (or) Explain for the
  #     training data.
  # @!attribute [rw] logging_sampling_strategy
  #   @return [::Google::Cloud::AIPlatform::V1::SamplingStrategy]
  #     Strategy to sample data from Training Dataset.
  #     If not set, we process the whole dataset.
  class TrainingDataset
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods
  end

  # The config for Training & Prediction data skew detection. It specifies the
  # training dataset sources and the skew detection parameters.
  # @!attribute [rw] skew_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. If a feature needs to
  #     be monitored for skew, a value threshold must be configured for that
  #     feature. The threshold here is against feature distribution distance
  #     between the training and prediction feature.
  # @!attribute [rw] attribution_score_skew_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. The threshold here is
  #     against attribution score distance between the training and prediction
  #     feature.
  # @!attribute [rw] default_skew_threshold
  #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
  #     Skew anomaly detection threshold used by all features.
  #     When the per-feature thresholds are not set, this field can be used to
  #     specify a threshold for all features.
  class TrainingPredictionSkewDetectionConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class SkewThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class AttributionScoreSkewThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end
  end

  # The config for Prediction data drift detection.
  # @!attribute [rw] drift_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. If a feature needs to
  #     be monitored for drift, a value threshold must be configured for that
  #     feature. The threshold here is against feature distribution distance
  #     between different time windws.
  # @!attribute [rw] attribution_score_drift_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. The threshold here is
  #     against attribution score distance between different time windows.
  # @!attribute [rw] default_drift_threshold
  #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
  #     Drift anomaly detection threshold used by all features.
  #     When the per-feature thresholds are not set, this field can be used to
  #     specify a threshold for all features.
  class PredictionDriftDetectionConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class DriftThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class AttributionScoreDriftThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end
  end

  # The config for integrating with Vertex Explainable AI. Only applicable if
  # the Model has explanation_spec populated.
  # @!attribute [rw] enable_feature_attributes
  #   @return [::Boolean]
  #     If want to analyze the Vertex Explainable AI feature attribute scores or
  #     not. If set to true, Vertex AI will log the feature attributions from
  #     explain response and do the skew/drift detection for them.
  # @!attribute [rw] explanation_baseline
  #   @return [::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::ExplanationConfig::ExplanationBaseline]
  #     Predictions generated by the BatchPredictionJob using baseline dataset.
  class ExplanationConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # Output from
    # {::Google::Cloud::AIPlatform::V1::BatchPredictionJob BatchPredictionJob} for
    # Model Monitoring baseline dataset, which can be used to generate baseline
    # attribution scores.
    # @!attribute [rw] gcs
    #   @return [::Google::Cloud::AIPlatform::V1::GcsDestination]
    #     Cloud Storage location for BatchExplain output.
    # @!attribute [rw] bigquery
    #   @return [::Google::Cloud::AIPlatform::V1::BigQueryDestination]
    #     BigQuery location for BatchExplain output.
    # @!attribute [rw] prediction_format
    #   @return [::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::ExplanationConfig::ExplanationBaseline::PredictionFormat]
    #     The storage format of the predictions generated BatchPrediction job.
    class ExplanationBaseline
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods

      # The storage format of the predictions generated BatchPrediction job.
      module PredictionFormat
        # Should not be set.
        PREDICTION_FORMAT_UNSPECIFIED = 0

        # Predictions are in JSONL files.
        JSONL = 2

        # Predictions are in BigQuery.
        BIGQUERY = 3
      end
    end
  end
end

#training_prediction_skew_detection_config ⇒ ::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::TrainingPredictionSkewDetectionConfig

Returns The config for skew between training data and prediction data.

Returns:



39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
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
 
# File 'proto_docs/google/cloud/aiplatform/v1/model_monitoring.rb', line 39

class ModelMonitoringObjectiveConfig
  include ::Google::Protobuf::MessageExts
  extend ::Google::Protobuf::MessageExts::ClassMethods

  # Training Dataset information.
  # @!attribute [rw] dataset
  #   @return [::String]
  #     The resource name of the Dataset used to train this Model.
  # @!attribute [rw] gcs_source
  #   @return [::Google::Cloud::AIPlatform::V1::GcsSource]
  #     The Google Cloud Storage uri of the unmanaged Dataset used to train
  #     this Model.
  # @!attribute [rw] bigquery_source
  #   @return [::Google::Cloud::AIPlatform::V1::BigQuerySource]
  #     The BigQuery table of the unmanaged Dataset used to train this
  #     Model.
  # @!attribute [rw] data_format
  #   @return [::String]
  #     Data format of the dataset, only applicable if the input is from
  #     Google Cloud Storage.
  #     The possible formats are:
  #
  #     "tf-record"
  #     The source file is a TFRecord file.
  #
  #     "csv"
  #     The source file is a CSV file.
  #     "jsonl"
  #     The source file is a JSONL file.
  # @!attribute [rw] target_field
  #   @return [::String]
  #     The target field name the model is to predict.
  #     This field will be excluded when doing Predict and (or) Explain for the
  #     training data.
  # @!attribute [rw] logging_sampling_strategy
  #   @return [::Google::Cloud::AIPlatform::V1::SamplingStrategy]
  #     Strategy to sample data from Training Dataset.
  #     If not set, we process the whole dataset.
  class TrainingDataset
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods
  end

  # The config for Training & Prediction data skew detection. It specifies the
  # training dataset sources and the skew detection parameters.
  # @!attribute [rw] skew_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. If a feature needs to
  #     be monitored for skew, a value threshold must be configured for that
  #     feature. The threshold here is against feature distribution distance
  #     between the training and prediction feature.
  # @!attribute [rw] attribution_score_skew_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. The threshold here is
  #     against attribution score distance between the training and prediction
  #     feature.
  # @!attribute [rw] default_skew_threshold
  #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
  #     Skew anomaly detection threshold used by all features.
  #     When the per-feature thresholds are not set, this field can be used to
  #     specify a threshold for all features.
  class TrainingPredictionSkewDetectionConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class SkewThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class AttributionScoreSkewThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end
  end

  # The config for Prediction data drift detection.
  # @!attribute [rw] drift_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. If a feature needs to
  #     be monitored for drift, a value threshold must be configured for that
  #     feature. The threshold here is against feature distribution distance
  #     between different time windws.
  # @!attribute [rw] attribution_score_drift_thresholds
  #   @return [::Google::Protobuf::Map{::String => ::Google::Cloud::AIPlatform::V1::ThresholdConfig}]
  #     Key is the feature name and value is the threshold. The threshold here is
  #     against attribution score distance between different time windows.
  # @!attribute [rw] default_drift_threshold
  #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
  #     Drift anomaly detection threshold used by all features.
  #     When the per-feature thresholds are not set, this field can be used to
  #     specify a threshold for all features.
  class PredictionDriftDetectionConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class DriftThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end

    # @!attribute [rw] key
    #   @return [::String]
    # @!attribute [rw] value
    #   @return [::Google::Cloud::AIPlatform::V1::ThresholdConfig]
    class AttributionScoreDriftThresholdsEntry
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods
    end
  end

  # The config for integrating with Vertex Explainable AI. Only applicable if
  # the Model has explanation_spec populated.
  # @!attribute [rw] enable_feature_attributes
  #   @return [::Boolean]
  #     If want to analyze the Vertex Explainable AI feature attribute scores or
  #     not. If set to true, Vertex AI will log the feature attributions from
  #     explain response and do the skew/drift detection for them.
  # @!attribute [rw] explanation_baseline
  #   @return [::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::ExplanationConfig::ExplanationBaseline]
  #     Predictions generated by the BatchPredictionJob using baseline dataset.
  class ExplanationConfig
    include ::Google::Protobuf::MessageExts
    extend ::Google::Protobuf::MessageExts::ClassMethods

    # Output from
    # {::Google::Cloud::AIPlatform::V1::BatchPredictionJob BatchPredictionJob} for
    # Model Monitoring baseline dataset, which can be used to generate baseline
    # attribution scores.
    # @!attribute [rw] gcs
    #   @return [::Google::Cloud::AIPlatform::V1::GcsDestination]
    #     Cloud Storage location for BatchExplain output.
    # @!attribute [rw] bigquery
    #   @return [::Google::Cloud::AIPlatform::V1::BigQueryDestination]
    #     BigQuery location for BatchExplain output.
    # @!attribute [rw] prediction_format
    #   @return [::Google::Cloud::AIPlatform::V1::ModelMonitoringObjectiveConfig::ExplanationConfig::ExplanationBaseline::PredictionFormat]
    #     The storage format of the predictions generated BatchPrediction job.
    class ExplanationBaseline
      include ::Google::Protobuf::MessageExts
      extend ::Google::Protobuf::MessageExts::ClassMethods

      # The storage format of the predictions generated BatchPrediction job.
      module PredictionFormat
        # Should not be set.
        PREDICTION_FORMAT_UNSPECIFIED = 0

        # Predictions are in JSONL files.
        JSONL = 2

        # Predictions are in BigQuery.
        BIGQUERY = 3
      end
    end
  end
end