var (
	Model_ModelType_name = map[int32]string{
		0:  "MODEL_TYPE_UNSPECIFIED",
		1:  "LINEAR_REGRESSION",
		2:  "LOGISTIC_REGRESSION",
		3:  "KMEANS",
		4:  "MATRIX_FACTORIZATION",
		5:  "DNN_CLASSIFIER",
		6:  "TENSORFLOW",
		7:  "DNN_REGRESSOR",
		9:  "BOOSTED_TREE_REGRESSOR",
		10: "BOOSTED_TREE_CLASSIFIER",
		11: "ARIMA",
		12: "AUTOML_REGRESSOR",
		13: "AUTOML_CLASSIFIER",
		19: "ARIMA_PLUS",
	}
	Model_ModelType_value = map[string]int32{
		"MODEL_TYPE_UNSPECIFIED":  0,
		"LINEAR_REGRESSION":       1,
		"LOGISTIC_REGRESSION":     2,
		"KMEANS":                  3,
		"MATRIX_FACTORIZATION":    4,
		"DNN_CLASSIFIER":          5,
		"TENSORFLOW":              6,
		"DNN_REGRESSOR":           7,
		"BOOSTED_TREE_REGRESSOR":  9,
		"BOOSTED_TREE_CLASSIFIER": 10,
		"ARIMA":                   11,
		"AUTOML_REGRESSOR":        12,
		"AUTOML_CLASSIFIER":       13,
		"ARIMA_PLUS":              19,
	}
)

var (
	Model_LossType_name = map[int32]string{
		0: "LOSS_TYPE_UNSPECIFIED",
		1: "MEAN_SQUARED_LOSS",
		2: "MEAN_LOG_LOSS",
	}
	Model_LossType_value = map[string]int32{
		"LOSS_TYPE_UNSPECIFIED": 0,
		"MEAN_SQUARED_LOSS":     1,
		"MEAN_LOG_LOSS":         2,
	}
)

var (
	Model_DistanceType_name = map[int32]string{
		0: "DISTANCE_TYPE_UNSPECIFIED",
		1: "EUCLIDEAN",
		2: "COSINE",
	}
	Model_DistanceType_value = map[string]int32{
		"DISTANCE_TYPE_UNSPECIFIED": 0,
		"EUCLIDEAN":                 1,
		"COSINE":                    2,
	}
)

var (
	Model_DataSplitMethod_name = map[int32]string{
		0: "DATA_SPLIT_METHOD_UNSPECIFIED",
		1: "RANDOM",
		2: "CUSTOM",
		3: "SEQUENTIAL",
		4: "NO_SPLIT",
		5: "AUTO_SPLIT",
	}
	Model_DataSplitMethod_value = map[string]int32{
		"DATA_SPLIT_METHOD_UNSPECIFIED": 0,
		"RANDOM":                        1,
		"CUSTOM":                        2,
		"SEQUENTIAL":                    3,
		"NO_SPLIT":                      4,
		"AUTO_SPLIT":                    5,
	}
)

var (
	Model_DataFrequency_name = map[int32]string{
		0: "DATA_FREQUENCY_UNSPECIFIED",
		1: "AUTO_FREQUENCY",
		2: "YEARLY",
		3: "QUARTERLY",
		4: "MONTHLY",
		5: "WEEKLY",
		6: "DAILY",
		7: "HOURLY",
		8: "PER_MINUTE",
	}
	Model_DataFrequency_value = map[string]int32{
		"DATA_FREQUENCY_UNSPECIFIED": 0,
		"AUTO_FREQUENCY":             1,
		"YEARLY":                     2,
		"QUARTERLY":                  3,
		"MONTHLY":                    4,
		"WEEKLY":                     5,
		"DAILY":                      6,
		"HOURLY":                     7,
		"PER_MINUTE":                 8,
	}
)

var (
	Model_HolidayRegion_name = map[int32]string{
		0:  "HOLIDAY_REGION_UNSPECIFIED",
		1:  "GLOBAL",
		2:  "NA",
		3:  "JAPAC",
		4:  "EMEA",
		5:  "LAC",
		6:  "AE",
		7:  "AR",
		8:  "AT",
		9:  "AU",
		10: "BE",
		11: "BR",
		12: "CA",
		13: "CH",
		14: "CL",
		15: "CN",
		16: "CO",
		17: "CS",
		18: "CZ",
		19: "DE",
		20: "DK",
		21: "DZ",
		22: "EC",
		23: "EE",
		24: "EG",
		25: "ES",
		26: "FI",
		27: "FR",
		28: "GB",
		29: "GR",
		30: "HK",
		31: "HU",
		32: "ID",
		33: "IE",
		34: "IL",
		35: "IN",
		36: "IR",
		37: "IT",
		38: "JP",
		39: "KR",
		40: "LV",
		41: "MA",
		42: "MX",
		43: "MY",
		44: "NG",
		45: "NL",
		46: "NO",
		47: "NZ",
		48: "PE",
		49: "PH",
		50: "PK",
		51: "PL",
		52: "PT",
		53: "RO",
		54: "RS",
		55: "RU",
		56: "SA",
		57: "SE",
		58: "SG",
		59: "SI",
		60: "SK",
		61: "TH",
		62: "TR",
		63: "TW",
		64: "UA",
		65: "US",
		66: "VE",
		67: "VN",
		68: "ZA",
	}
	Model_HolidayRegion_value = map[string]int32{
		"HOLIDAY_REGION_UNSPECIFIED": 0,
		"GLOBAL":                     1,
		"NA":                         2,
		"JAPAC":                      3,
		"EMEA":                       4,
		"LAC":                        5,
		"AE":                         6,
		"AR":                         7,
		"AT":                         8,
		"AU":                         9,
		"BE":                         10,
		"BR":                         11,
		"CA":                         12,
		"CH":                         13,
		"CL":                         14,
		"CN":                         15,
		"CO":                         16,
		"CS":                         17,
		"CZ":                         18,
		"DE":                         19,
		"DK":                         20,
		"DZ":                         21,
		"EC":                         22,
		"EE":                         23,
		"EG":                         24,
		"ES":                         25,
		"FI":                         26,
		"FR":                         27,
		"GB":                         28,
		"GR":                         29,
		"HK":                         30,
		"HU":                         31,
		"ID":                         32,
		"IE":                         33,
		"IL":                         34,
		"IN":                         35,
		"IR":                         36,
		"IT":                         37,
		"JP":                         38,
		"KR":                         39,
		"LV":                         40,
		"MA":                         41,
		"MX":                         42,
		"MY":                         43,
		"NG":                         44,
		"NL":                         45,
		"NO":                         46,
		"NZ":                         47,
		"PE":                         48,
		"PH":                         49,
		"PK":                         50,
		"PL":                         51,
		"PT":                         52,
		"RO":                         53,
		"RS":                         54,
		"RU":                         55,
		"SA":                         56,
		"SE":                         57,
		"SG":                         58,
		"SI":                         59,
		"SK":                         60,
		"TH":                         61,
		"TR":                         62,
		"TW":                         63,
		"UA":                         64,
		"US":                         65,
		"VE":                         66,
		"VN":                         67,
		"ZA":                         68,
	}
)

var (
	Model_LearnRateStrategy_name = map[int32]string{
		0: "LEARN_RATE_STRATEGY_UNSPECIFIED",
		1: "LINE_SEARCH",
		2: "CONSTANT",
	}
	Model_LearnRateStrategy_value = map[string]int32{
		"LEARN_RATE_STRATEGY_UNSPECIFIED": 0,
		"LINE_SEARCH":                     1,
		"CONSTANT":                        2,
	}
)

var (
	Model_OptimizationStrategy_name = map[int32]string{
		0: "OPTIMIZATION_STRATEGY_UNSPECIFIED",
		1: "BATCH_GRADIENT_DESCENT",
		2: "NORMAL_EQUATION",
	}
	Model_OptimizationStrategy_value = map[string]int32{
		"OPTIMIZATION_STRATEGY_UNSPECIFIED": 0,
		"BATCH_GRADIENT_DESCENT":            1,
		"NORMAL_EQUATION":                   2,
	}
)

var (
	Model_FeedbackType_name = map[int32]string{
		0: "FEEDBACK_TYPE_UNSPECIFIED",
		1: "IMPLICIT",
		2: "EXPLICIT",
	}
	Model_FeedbackType_value = map[string]int32{
		"FEEDBACK_TYPE_UNSPECIFIED": 0,
		"IMPLICIT":                  1,
		"EXPLICIT":                  2,
	}
)

var (
	Model_SeasonalPeriod_SeasonalPeriodType_name = map[int32]string{
		0: "SEASONAL_PERIOD_TYPE_UNSPECIFIED",
		1: "NO_SEASONALITY",
		2: "DAILY",
		3: "WEEKLY",
		4: "MONTHLY",
		5: "QUARTERLY",
		6: "YEARLY",
	}
	Model_SeasonalPeriod_SeasonalPeriodType_value = map[string]int32{
		"SEASONAL_PERIOD_TYPE_UNSPECIFIED": 0,
		"NO_SEASONALITY":                   1,
		"DAILY":                            2,
		"WEEKLY":                           3,
		"MONTHLY":                          4,
		"QUARTERLY":                        5,
		"YEARLY":                           6,
	}
)

var (
	Model_KmeansEnums_KmeansInitializationMethod_name = map[int32]string{
		0: "KMEANS_INITIALIZATION_METHOD_UNSPECIFIED",
		1: "RANDOM",
		2: "CUSTOM",
		3: "KMEANS_PLUS_PLUS",
	}
	Model_KmeansEnums_KmeansInitializationMethod_value = map[string]int32{
		"KMEANS_INITIALIZATION_METHOD_UNSPECIFIED": 0,
		"RANDOM":           1,
		"CUSTOM":           2,
		"KMEANS_PLUS_PLUS": 3,
	}
)

var (
	StandardSqlDataType_TypeKind_name = map[int32]string{
		0:  "TYPE_KIND_UNSPECIFIED",
		2:  "INT64",
		5:  "BOOL",
		7:  "FLOAT64",
		8:  "STRING",
		9:  "BYTES",
		19: "TIMESTAMP",
		10: "DATE",
		20: "TIME",
		21: "DATETIME",
		26: "INTERVAL",
		22: "GEOGRAPHY",
		23: "NUMERIC",
		24: "BIGNUMERIC",
		25: "JSON",
		16: "ARRAY",
		17: "STRUCT",
	}
	StandardSqlDataType_TypeKind_value = map[string]int32{
		"TYPE_KIND_UNSPECIFIED": 0,
		"INT64":                 2,
		"BOOL":                  5,
		"FLOAT64":               7,
		"STRING":                8,
		"BYTES":                 9,
		"TIMESTAMP":             19,
		"DATE":                  10,
		"TIME":                  20,
		"DATETIME":              21,
		"INTERVAL":              26,
		"GEOGRAPHY":             22,
		"NUMERIC":               23,
		"BIGNUMERIC":            24,
		"JSON":                  25,
		"ARRAY":                 16,
		"STRUCT":                17,
	}
)

var File_google_cloud_bigquery_v2_encryption_config_proto protoreflect.FileDescriptor

var File_google_cloud_bigquery_v2_model_proto protoreflect.FileDescriptor

var File_google_cloud_bigquery_v2_model_reference_proto protoreflect.FileDescriptor

var File_google_cloud_bigquery_v2_standard_sql_proto protoreflect.FileDescriptor

var File_google_cloud_bigquery_v2_table_reference_proto protoreflect.FileDescriptor

type DeleteModelRequest struct {

	// Required. Project ID of the model to delete.
	ProjectId string `protobuf:"bytes,1,opt,name=project_id,json=projectId,proto3" json:"project_id,omitempty"`
	// Required. Dataset ID of the model to delete.
	DatasetId string `protobuf:"bytes,2,opt,name=dataset_id,json=datasetId,proto3" json:"dataset_id,omitempty"`
	// Required. Model ID of the model to delete.
	ModelId string `protobuf:"bytes,3,opt,name=model_id,json=modelId,proto3" json:"model_id,omitempty"`
	// contains filtered or unexported fields
}

type EncryptionConfiguration struct {

	// Optional. Describes the Cloud KMS encryption key that will be used to
	// protect destination BigQuery table. The BigQuery Service Account associated
	// with your project requires access to this encryption key.
	KmsKeyName *wrapperspb.StringValue `protobuf:"bytes,1,opt,name=kms_key_name,json=kmsKeyName,proto3" json:"kms_key_name,omitempty"`
	// contains filtered or unexported fields
}

type GetModelRequest struct {

	// Required. Project ID of the requested model.
	ProjectId string `protobuf:"bytes,1,opt,name=project_id,json=projectId,proto3" json:"project_id,omitempty"`
	// Required. Dataset ID of the requested model.
	DatasetId string `protobuf:"bytes,2,opt,name=dataset_id,json=datasetId,proto3" json:"dataset_id,omitempty"`
	// Required. Model ID of the requested model.
	ModelId string `protobuf:"bytes,3,opt,name=model_id,json=modelId,proto3" json:"model_id,omitempty"`
	// contains filtered or unexported fields
}

type ListModelsRequest struct {

	// Required. Project ID of the models to list.
	ProjectId string `protobuf:"bytes,1,opt,name=project_id,json=projectId,proto3" json:"project_id,omitempty"`
	// Required. Dataset ID of the models to list.
	DatasetId string `protobuf:"bytes,2,opt,name=dataset_id,json=datasetId,proto3" json:"dataset_id,omitempty"`
	// The maximum number of results to return in a single response page.
	// Leverage the page tokens to iterate through the entire collection.
	MaxResults *wrapperspb.UInt32Value `protobuf:"bytes,3,opt,name=max_results,json=maxResults,proto3" json:"max_results,omitempty"`
	// Page token, returned by a previous call to request the next page of
	// results
	PageToken string `protobuf:"bytes,4,opt,name=page_token,json=pageToken,proto3" json:"page_token,omitempty"`
	// contains filtered or unexported fields
}

type ListModelsResponse struct {

	// Models in the requested dataset. Only the following fields are populated:
	// model_reference, model_type, creation_time, last_modified_time and
	// labels.
	Models []*Model `protobuf:"bytes,1,rep,name=models,proto3" json:"models,omitempty"`
	// A token to request the next page of results.
	NextPageToken string `protobuf:"bytes,2,opt,name=next_page_token,json=nextPageToken,proto3" json:"next_page_token,omitempty"`
	// contains filtered or unexported fields
}

type Model struct {

	// Output only. A hash of this resource.
	Etag string `protobuf:"bytes,1,opt,name=etag,proto3" json:"etag,omitempty"`
	// Required. Unique identifier for this model.
	ModelReference *ModelReference `protobuf:"bytes,2,opt,name=model_reference,json=modelReference,proto3" json:"model_reference,omitempty"`
	// Output only. The time when this model was created, in millisecs since the epoch.
	CreationTime int64 `protobuf:"varint,5,opt,name=creation_time,json=creationTime,proto3" json:"creation_time,omitempty"`
	// Output only. The time when this model was last modified, in millisecs since the epoch.
	LastModifiedTime int64 `protobuf:"varint,6,opt,name=last_modified_time,json=lastModifiedTime,proto3" json:"last_modified_time,omitempty"`
	// Optional. A user-friendly description of this model.
	Description string `protobuf:"bytes,12,opt,name=description,proto3" json:"description,omitempty"`
	// Optional. A descriptive name for this model.
	FriendlyName string `protobuf:"bytes,14,opt,name=friendly_name,json=friendlyName,proto3" json:"friendly_name,omitempty"`
	// The labels associated with this model. You can use these to organize
	// and group your models. Label keys and values can be no longer
	// than 63 characters, can only contain lowercase letters, numeric
	// characters, underscores and dashes. International characters are allowed.
	// Label values are optional. Label keys must start with a letter and each
	// label in the list must have a different key.
	Labels map[string]string `protobuf:"bytes,15,rep,name=labels,proto3" json:"labels,omitempty" protobuf_key:"bytes,1,opt,name=key,proto3" protobuf_val:"bytes,2,opt,name=value,proto3"`
	// Optional. The time when this model expires, in milliseconds since the epoch.
	// If not present, the model will persist indefinitely. Expired models
	// will be deleted and their storage reclaimed.  The defaultTableExpirationMs
	// property of the encapsulating dataset can be used to set a default
	// expirationTime on newly created models.
	ExpirationTime int64 `protobuf:"varint,16,opt,name=expiration_time,json=expirationTime,proto3" json:"expiration_time,omitempty"`
	// Output only. The geographic location where the model resides. This value
	// is inherited from the dataset.
	Location string `protobuf:"bytes,13,opt,name=location,proto3" json:"location,omitempty"`
	// Custom encryption configuration (e.g., Cloud KMS keys). This shows the
	// encryption configuration of the model data while stored in BigQuery
	// storage. This field can be used with PatchModel to update encryption key
	// for an already encrypted model.
	EncryptionConfiguration *EncryptionConfiguration `protobuf:"bytes,17,opt,name=encryption_configuration,json=encryptionConfiguration,proto3" json:"encryption_configuration,omitempty"`
	// Output only. Type of the model resource.
	ModelType Model_ModelType `protobuf:"varint,7,opt,name=model_type,json=modelType,proto3,enum=google.cloud.bigquery.v2.Model_ModelType" json:"model_type,omitempty"`
	// Output only. Information for all training runs in increasing order of start_time.
	TrainingRuns []*Model_TrainingRun `protobuf:"bytes,9,rep,name=training_runs,json=trainingRuns,proto3" json:"training_runs,omitempty"`
	// Output only. Input feature columns that were used to train this model.
	FeatureColumns []*StandardSqlField `protobuf:"bytes,10,rep,name=feature_columns,json=featureColumns,proto3" json:"feature_columns,omitempty"`
	// Output only. Label columns that were used to train this model.
	// The output of the model will have a "predicted_" prefix to these columns.
	LabelColumns []*StandardSqlField `protobuf:"bytes,11,rep,name=label_columns,json=labelColumns,proto3" json:"label_columns,omitempty"`
	// The best trial_id across all training runs.
	//
	// Deprecated: Do not use.
	BestTrialId int64 `protobuf:"varint,19,opt,name=best_trial_id,json=bestTrialId,proto3" json:"best_trial_id,omitempty"`
	// contains filtered or unexported fields
}

type ModelReference struct {

	// Required. The ID of the project containing this model.
	ProjectId string `protobuf:"bytes,1,opt,name=project_id,json=projectId,proto3" json:"project_id,omitempty"`
	// Required. The ID of the dataset containing this model.
	DatasetId string `protobuf:"bytes,2,opt,name=dataset_id,json=datasetId,proto3" json:"dataset_id,omitempty"`
	// Required. The ID of the model. The ID must contain only
	// letters (a-z, A-Z), numbers (0-9), or underscores (_). The maximum
	// length is 1,024 characters.
	ModelId string `protobuf:"bytes,3,opt,name=model_id,json=modelId,proto3" json:"model_id,omitempty"`
	// contains filtered or unexported fields
}

type ModelServiceClient interface {
	// Gets the specified model resource by model ID.
	GetModel(ctx context.Context, in *GetModelRequest, opts ...grpc.CallOption) (*Model, error)
	// Lists all models in the specified dataset. Requires the READER dataset
	// role. After retrieving the list of models, you can get information about a
	// particular model by calling the models.get method.
	ListModels(ctx context.Context, in *ListModelsRequest, opts ...grpc.CallOption) (*ListModelsResponse, error)
	// Patch specific fields in the specified model.
	PatchModel(ctx context.Context, in *PatchModelRequest, opts ...grpc.CallOption) (*Model, error)
	// Deletes the model specified by modelId from the dataset.
	DeleteModel(ctx context.Context, in *DeleteModelRequest, opts ...grpc.CallOption) (*emptypb.Empty, error)
}

type ModelServiceServer interface {
	// Gets the specified model resource by model ID.
	GetModel(context.Context, *GetModelRequest) (*Model, error)
	// Lists all models in the specified dataset. Requires the READER dataset
	// role. After retrieving the list of models, you can get information about a
	// particular model by calling the models.get method.
	ListModels(context.Context, *ListModelsRequest) (*ListModelsResponse, error)
	// Patch specific fields in the specified model.
	PatchModel(context.Context, *PatchModelRequest) (*Model, error)
	// Deletes the model specified by modelId from the dataset.
	DeleteModel(context.Context, *DeleteModelRequest) (*emptypb.Empty, error)
}

type Model_AggregateClassificationMetrics struct {

	// Precision is the fraction of actual positive predictions that had
	// positive actual labels. For multiclass this is a macro-averaged
	// metric treating each class as a binary classifier.
	Precision *wrapperspb.DoubleValue `protobuf:"bytes,1,opt,name=precision,proto3" json:"precision,omitempty"`
	// Recall is the fraction of actual positive labels that were given a
	// positive prediction. For multiclass this is a macro-averaged metric.
	Recall *wrapperspb.DoubleValue `protobuf:"bytes,2,opt,name=recall,proto3" json:"recall,omitempty"`
	// Accuracy is the fraction of predictions given the correct label. For
	// multiclass this is a micro-averaged metric.
	Accuracy *wrapperspb.DoubleValue `protobuf:"bytes,3,opt,name=accuracy,proto3" json:"accuracy,omitempty"`
	// Threshold at which the metrics are computed. For binary
	// classification models this is the positive class threshold.
	// For multi-class classfication models this is the confidence
	// threshold.
	Threshold *wrapperspb.DoubleValue `protobuf:"bytes,4,opt,name=threshold,proto3" json:"threshold,omitempty"`
	// The F1 score is an average of recall and precision. For multiclass
	// this is a macro-averaged metric.
	F1Score *wrapperspb.DoubleValue `protobuf:"bytes,5,opt,name=f1_score,json=f1Score,proto3" json:"f1_score,omitempty"`
	// Logarithmic Loss. For multiclass this is a macro-averaged metric.
	LogLoss *wrapperspb.DoubleValue `protobuf:"bytes,6,opt,name=log_loss,json=logLoss,proto3" json:"log_loss,omitempty"`
	// Area Under a ROC Curve. For multiclass this is a macro-averaged
	// metric.
	RocAuc *wrapperspb.DoubleValue `protobuf:"bytes,7,opt,name=roc_auc,json=rocAuc,proto3" json:"roc_auc,omitempty"`
	// contains filtered or unexported fields
}

type Model_ArimaFittingMetrics struct {

	// Log-likelihood.
	LogLikelihood float64 `protobuf:"fixed64,1,opt,name=log_likelihood,json=logLikelihood,proto3" json:"log_likelihood,omitempty"`
	// AIC.
	Aic float64 `protobuf:"fixed64,2,opt,name=aic,proto3" json:"aic,omitempty"`
	// Variance.
	Variance float64 `protobuf:"fixed64,3,opt,name=variance,proto3" json:"variance,omitempty"`
	// contains filtered or unexported fields
}

type Model_ArimaForecastingMetrics struct {

	// Non-seasonal order.
	//
	// Deprecated: Do not use.
	NonSeasonalOrder []*Model_ArimaOrder `protobuf:"bytes,1,rep,name=non_seasonal_order,json=nonSeasonalOrder,proto3" json:"non_seasonal_order,omitempty"`
	// Arima model fitting metrics.
	//
	// Deprecated: Do not use.
	ArimaFittingMetrics []*Model_ArimaFittingMetrics `protobuf:"bytes,2,rep,name=arima_fitting_metrics,json=arimaFittingMetrics,proto3" json:"arima_fitting_metrics,omitempty"`
	// Seasonal periods. Repeated because multiple periods are supported for one
	// time series.
	//
	// Deprecated: Do not use.
	SeasonalPeriods []Model_SeasonalPeriod_SeasonalPeriodType `protobuf:"varint,3,rep,packed,name=seasonal_periods,json=seasonalPeriods,proto3,enum=google.cloud.bigquery.v2.Model_SeasonalPeriod_SeasonalPeriodType" json:"seasonal_periods,omitempty"`
	// Whether Arima model fitted with drift or not. It is always false when d
	// is not 1.
	//
	// Deprecated: Do not use.
	HasDrift []bool `protobuf:"varint,4,rep,packed,name=has_drift,json=hasDrift,proto3" json:"has_drift,omitempty"`
	// Id to differentiate different time series for the large-scale case.
	//
	// Deprecated: Do not use.
	TimeSeriesId []string `protobuf:"bytes,5,rep,name=time_series_id,json=timeSeriesId,proto3" json:"time_series_id,omitempty"`
	// Repeated as there can be many metric sets (one for each model) in
	// auto-arima and the large-scale case.
	ArimaSingleModelForecastingMetrics []*Model_ArimaForecastingMetrics_ArimaSingleModelForecastingMetrics `protobuf:"bytes,6,rep,name=arima_single_model_forecasting_metrics,json=arimaSingleModelForecastingMetrics,proto3" json:"arima_single_model_forecasting_metrics,omitempty"`
	// contains filtered or unexported fields
}

type Model_ArimaForecastingMetrics_ArimaSingleModelForecastingMetrics struct {

	// Non-seasonal order.
	NonSeasonalOrder *Model_ArimaOrder `protobuf:"bytes,1,opt,name=non_seasonal_order,json=nonSeasonalOrder,proto3" json:"non_seasonal_order,omitempty"`
	// Arima fitting metrics.
	ArimaFittingMetrics *Model_ArimaFittingMetrics `protobuf:"bytes,2,opt,name=arima_fitting_metrics,json=arimaFittingMetrics,proto3" json:"arima_fitting_metrics,omitempty"`
	// Is arima model fitted with drift or not. It is always false when d
	// is not 1.
	HasDrift bool `protobuf:"varint,3,opt,name=has_drift,json=hasDrift,proto3" json:"has_drift,omitempty"`
	// The time_series_id value for this time series. It will be one of
	// the unique values from the time_series_id_column specified during
	// ARIMA model training. Only present when time_series_id_column
	// training option was used.
	TimeSeriesId string `protobuf:"bytes,4,opt,name=time_series_id,json=timeSeriesId,proto3" json:"time_series_id,omitempty"`
	// The tuple of time_series_ids identifying this time series. It will
	// be one of the unique tuples of values present in the
	// time_series_id_columns specified during ARIMA model training. Only
	// present when time_series_id_columns training option was used and
	// the order of values here are same as the order of
	// time_series_id_columns.
	TimeSeriesIds []string `protobuf:"bytes,9,rep,name=time_series_ids,json=timeSeriesIds,proto3" json:"time_series_ids,omitempty"`
	// Seasonal periods. Repeated because multiple periods are supported
	// for one time series.
	SeasonalPeriods []Model_SeasonalPeriod_SeasonalPeriodType `protobuf:"varint,5,rep,packed,name=seasonal_periods,json=seasonalPeriods,proto3,enum=google.cloud.bigquery.v2.Model_SeasonalPeriod_SeasonalPeriodType" json:"seasonal_periods,omitempty"`
	// If true, holiday_effect is a part of time series decomposition result.
	HasHolidayEffect *wrapperspb.BoolValue `protobuf:"bytes,6,opt,name=has_holiday_effect,json=hasHolidayEffect,proto3" json:"has_holiday_effect,omitempty"`
	// If true, spikes_and_dips is a part of time series decomposition result.
	HasSpikesAndDips *wrapperspb.BoolValue `protobuf:"bytes,7,opt,name=has_spikes_and_dips,json=hasSpikesAndDips,proto3" json:"has_spikes_and_dips,omitempty"`
	// If true, step_changes is a part of time series decomposition result.
	HasStepChanges *wrapperspb.BoolValue `protobuf:"bytes,8,opt,name=has_step_changes,json=hasStepChanges,proto3" json:"has_step_changes,omitempty"`
	// contains filtered or unexported fields
}

type Model_ArimaOrder struct {

	// Order of the autoregressive part.
	P int64 `protobuf:"varint,1,opt,name=p,proto3" json:"p,omitempty"`
	// Order of the differencing part.
	D int64 `protobuf:"varint,2,opt,name=d,proto3" json:"d,omitempty"`
	// Order of the moving-average part.
	Q int64 `protobuf:"varint,3,opt,name=q,proto3" json:"q,omitempty"`
	// contains filtered or unexported fields
}

type Model_BinaryClassificationMetrics struct {

	// Aggregate classification metrics.
	AggregateClassificationMetrics *Model_AggregateClassificationMetrics `protobuf:"bytes,1,opt,name=aggregate_classification_metrics,json=aggregateClassificationMetrics,proto3" json:"aggregate_classification_metrics,omitempty"`
	// Binary confusion matrix at multiple thresholds.
	BinaryConfusionMatrixList []*Model_BinaryClassificationMetrics_BinaryConfusionMatrix `protobuf:"bytes,2,rep,name=binary_confusion_matrix_list,json=binaryConfusionMatrixList,proto3" json:"binary_confusion_matrix_list,omitempty"`
	// Label representing the positive class.
	PositiveLabel string `protobuf:"bytes,3,opt,name=positive_label,json=positiveLabel,proto3" json:"positive_label,omitempty"`
	// Label representing the negative class.
	NegativeLabel string `protobuf:"bytes,4,opt,name=negative_label,json=negativeLabel,proto3" json:"negative_label,omitempty"`
	// contains filtered or unexported fields
}

type Model_BinaryClassificationMetrics_BinaryConfusionMatrix struct {

	// Threshold value used when computing each of the following metric.
	PositiveClassThreshold *wrapperspb.DoubleValue `protobuf:"bytes,1,opt,name=positive_class_threshold,json=positiveClassThreshold,proto3" json:"positive_class_threshold,omitempty"`
	// Number of true samples predicted as true.
	TruePositives *wrapperspb.Int64Value `protobuf:"bytes,2,opt,name=true_positives,json=truePositives,proto3" json:"true_positives,omitempty"`
	// Number of false samples predicted as true.
	FalsePositives *wrapperspb.Int64Value `protobuf:"bytes,3,opt,name=false_positives,json=falsePositives,proto3" json:"false_positives,omitempty"`
	// Number of true samples predicted as false.
	TrueNegatives *wrapperspb.Int64Value `protobuf:"bytes,4,opt,name=true_negatives,json=trueNegatives,proto3" json:"true_negatives,omitempty"`
	// Number of false samples predicted as false.
	FalseNegatives *wrapperspb.Int64Value `protobuf:"bytes,5,opt,name=false_negatives,json=falseNegatives,proto3" json:"false_negatives,omitempty"`
	// The fraction of actual positive predictions that had positive actual
	// labels.
	Precision *wrapperspb.DoubleValue `protobuf:"bytes,6,opt,name=precision,proto3" json:"precision,omitempty"`
	// The fraction of actual positive labels that were given a positive
	// prediction.
	Recall *wrapperspb.DoubleValue `protobuf:"bytes,7,opt,name=recall,proto3" json:"recall,omitempty"`
	// The equally weighted average of recall and precision.
	F1Score *wrapperspb.DoubleValue `protobuf:"bytes,8,opt,name=f1_score,json=f1Score,proto3" json:"f1_score,omitempty"`
	// The fraction of predictions given the correct label.
	Accuracy *wrapperspb.DoubleValue `protobuf:"bytes,9,opt,name=accuracy,proto3" json:"accuracy,omitempty"`
	// contains filtered or unexported fields
}

type Model_ClusteringMetrics struct {

	// Davies-Bouldin index.
	DaviesBouldinIndex *wrapperspb.DoubleValue `protobuf:"bytes,1,opt,name=davies_bouldin_index,json=daviesBouldinIndex,proto3" json:"davies_bouldin_index,omitempty"`
	// Mean of squared distances between each sample to its cluster centroid.
	MeanSquaredDistance *wrapperspb.DoubleValue `protobuf:"bytes,2,opt,name=mean_squared_distance,json=meanSquaredDistance,proto3" json:"mean_squared_distance,omitempty"`
	// Information for all clusters.
	Clusters []*Model_ClusteringMetrics_Cluster `protobuf:"bytes,3,rep,name=clusters,proto3" json:"clusters,omitempty"`
	// contains filtered or unexported fields
}

type Model_ClusteringMetrics_Cluster struct {

	// Centroid id.
	CentroidId int64 `protobuf:"varint,1,opt,name=centroid_id,json=centroidId,proto3" json:"centroid_id,omitempty"`
	// Values of highly variant features for this cluster.
	FeatureValues []*Model_ClusteringMetrics_Cluster_FeatureValue `protobuf:"bytes,2,rep,name=feature_values,json=featureValues,proto3" json:"feature_values,omitempty"`
	// Count of training data rows that were assigned to this cluster.
	Count *wrapperspb.Int64Value `protobuf:"bytes,3,opt,name=count,proto3" json:"count,omitempty"`
	// contains filtered or unexported fields
}

type Model_ClusteringMetrics_Cluster_FeatureValue struct {

	// The feature column name.
	FeatureColumn string `protobuf:"bytes,1,opt,name=feature_column,json=featureColumn,proto3" json:"feature_column,omitempty"`
	// Types that are assignable to Value:
	//	*Model_ClusteringMetrics_Cluster_FeatureValue_NumericalValue
	//	*Model_ClusteringMetrics_Cluster_FeatureValue_CategoricalValue_
	Value isModel_ClusteringMetrics_Cluster_FeatureValue_Value `protobuf_oneof:"value"`
	// contains filtered or unexported fields
}

type Model_ClusteringMetrics_Cluster_FeatureValue_CategoricalValue struct {

	// Counts of all categories for the categorical feature. If there are
	// more than ten categories, we return top ten (by count) and return
	// one more CategoryCount with category "_OTHER_" and count as
	// aggregate counts of remaining categories.
	CategoryCounts []*Model_ClusteringMetrics_Cluster_FeatureValue_CategoricalValue_CategoryCount `protobuf:"bytes,1,rep,name=category_counts,json=categoryCounts,proto3" json:"category_counts,omitempty"`
	// contains filtered or unexported fields
}

type Model_ClusteringMetrics_Cluster_FeatureValue_CategoricalValue_ struct {
	// The categorical feature value.
	CategoricalValue *Model_ClusteringMetrics_Cluster_FeatureValue_CategoricalValue `protobuf:"bytes,3,opt,name=categorical_value,json=categoricalValue,proto3,oneof"`
}

type Model_ClusteringMetrics_Cluster_FeatureValue_CategoricalValue_CategoryCount struct {

	// The name of category.
	Category string `protobuf:"bytes,1,opt,name=category,proto3" json:"category,omitempty"`
	// The count of training samples matching the category within the
	// cluster.
	Count *wrapperspb.Int64Value `protobuf:"bytes,2,opt,name=count,proto3" json:"count,omitempty"`
	// contains filtered or unexported fields
}

type Model_ClusteringMetrics_Cluster_FeatureValue_NumericalValue struct {
	// The numerical feature value. This is the centroid value for this
	// feature.
	NumericalValue *wrapperspb.DoubleValue `protobuf:"bytes,2,opt,name=numerical_value,json=numericalValue,proto3,oneof"`
}

type Model_DataFrequency int32

const (
	Model_DATA_FREQUENCY_UNSPECIFIED Model_DataFrequency = 0
	// Automatically inferred from timestamps.
	Model_AUTO_FREQUENCY Model_DataFrequency = 1
	// Yearly data.
	Model_YEARLY Model_DataFrequency = 2
	// Quarterly data.
	Model_QUARTERLY Model_DataFrequency = 3
	// Monthly data.
	Model_MONTHLY Model_DataFrequency = 4
	// Weekly data.
	Model_WEEKLY Model_DataFrequency = 5
	// Daily data.
	Model_DAILY Model_DataFrequency = 6
	// Hourly data.
	Model_HOURLY Model_DataFrequency = 7
	// Per-minute data.
	Model_PER_MINUTE Model_DataFrequency = 8
)

type Model_DataSplitMethod int32

const (
	Model_DATA_SPLIT_METHOD_UNSPECIFIED Model_DataSplitMethod = 0
	// Splits data randomly.
	Model_RANDOM Model_DataSplitMethod = 1
	// Splits data with the user provided tags.
	Model_CUSTOM Model_DataSplitMethod = 2
	// Splits data sequentially.
	Model_SEQUENTIAL Model_DataSplitMethod = 3
	// Data split will be skipped.
	Model_NO_SPLIT Model_DataSplitMethod = 4
	// Splits data automatically: Uses NO_SPLIT if the data size is small.
	// Otherwise uses RANDOM.
	Model_AUTO_SPLIT Model_DataSplitMethod = 5
)

type Model_DataSplitResult struct {

	// Table reference of the training data after split.
	TrainingTable *TableReference `protobuf:"bytes,1,opt,name=training_table,json=trainingTable,proto3" json:"training_table,omitempty"`
	// Table reference of the evaluation data after split.
	EvaluationTable *TableReference `protobuf:"bytes,2,opt,name=evaluation_table,json=evaluationTable,proto3" json:"evaluation_table,omitempty"`
	// contains filtered or unexported fields
}

type Model_DistanceType int32

const (
	Model_DISTANCE_TYPE_UNSPECIFIED Model_DistanceType = 0
	// Eculidean distance.
	Model_EUCLIDEAN Model_DistanceType = 1
	// Cosine distance.
	Model_COSINE Model_DistanceType = 2
)

type Model_EvaluationMetrics struct {

	// Types that are assignable to Metrics:
	//	*Model_EvaluationMetrics_RegressionMetrics
	//	*Model_EvaluationMetrics_BinaryClassificationMetrics
	//	*Model_EvaluationMetrics_MultiClassClassificationMetrics
	//	*Model_EvaluationMetrics_ClusteringMetrics
	//	*Model_EvaluationMetrics_RankingMetrics
	//	*Model_EvaluationMetrics_ArimaForecastingMetrics
	Metrics isModel_EvaluationMetrics_Metrics `protobuf_oneof:"metrics"`
	// contains filtered or unexported fields
}

type Model_EvaluationMetrics_ArimaForecastingMetrics struct {
	// Populated for ARIMA models.
	ArimaForecastingMetrics *Model_ArimaForecastingMetrics `protobuf:"bytes,6,opt,name=arima_forecasting_metrics,json=arimaForecastingMetrics,proto3,oneof"`
}

type Model_EvaluationMetrics_BinaryClassificationMetrics struct {
	// Populated for binary classification/classifier models.
	BinaryClassificationMetrics *Model_BinaryClassificationMetrics `protobuf:"bytes,2,opt,name=binary_classification_metrics,json=binaryClassificationMetrics,proto3,oneof"`
}

type Model_EvaluationMetrics_ClusteringMetrics struct {
	// Populated for clustering models.
	ClusteringMetrics *Model_ClusteringMetrics `protobuf:"bytes,4,opt,name=clustering_metrics,json=clusteringMetrics,proto3,oneof"`
}

type Model_EvaluationMetrics_MultiClassClassificationMetrics struct {
	// Populated for multi-class classification/classifier models.
	MultiClassClassificationMetrics *Model_MultiClassClassificationMetrics `protobuf:"bytes,3,opt,name=multi_class_classification_metrics,json=multiClassClassificationMetrics,proto3,oneof"`
}

type Model_EvaluationMetrics_RankingMetrics struct {
	// Populated for implicit feedback type matrix factorization models.
	RankingMetrics *Model_RankingMetrics `protobuf:"bytes,5,opt,name=ranking_metrics,json=rankingMetrics,proto3,oneof"`
}

type Model_EvaluationMetrics_RegressionMetrics struct {
	// Populated for regression models and explicit feedback type matrix
	// factorization models.
	RegressionMetrics *Model_RegressionMetrics `protobuf:"bytes,1,opt,name=regression_metrics,json=regressionMetrics,proto3,oneof"`
}

type Model_FeedbackType int32

const (
	Model_FEEDBACK_TYPE_UNSPECIFIED Model_FeedbackType = 0
	// Use weighted-als for implicit feedback problems.
	Model_IMPLICIT Model_FeedbackType = 1
	// Use nonweighted-als for explicit feedback problems.
	Model_EXPLICIT Model_FeedbackType = 2
)

type Model_GlobalExplanation struct {

	// A list of the top global explanations. Sorted by absolute value of
	// attribution in descending order.
	Explanations []*Model_GlobalExplanation_Explanation `protobuf:"bytes,1,rep,name=explanations,proto3" json:"explanations,omitempty"`
	// Class label for this set of global explanations. Will be empty/null for
	// binary logistic and linear regression models. Sorted alphabetically in
	// descending order.
	ClassLabel string `protobuf:"bytes,2,opt,name=class_label,json=classLabel,proto3" json:"class_label,omitempty"`
	// contains filtered or unexported fields
}

type Model_GlobalExplanation_Explanation struct {

	// Full name of the feature. For non-numerical features, will be
	// formatted like <column_name>.<encoded_feature_name>. Overall size of
	// feature name will always be truncated to first 120 characters.
	FeatureName string `protobuf:"bytes,1,opt,name=feature_name,json=featureName,proto3" json:"feature_name,omitempty"`
	// Attribution of feature.
	Attribution *wrapperspb.DoubleValue `protobuf:"bytes,2,opt,name=attribution,proto3" json:"attribution,omitempty"`
	// contains filtered or unexported fields
}

type Model_HolidayRegion int32

const (
	// Holiday region unspecified.
	Model_HOLIDAY_REGION_UNSPECIFIED Model_HolidayRegion = 0
	// Global.
	Model_GLOBAL Model_HolidayRegion = 1
	// North America.
	Model_NA Model_HolidayRegion = 2
	// Japan and Asia Pacific: Korea, Greater China, India, Australia, and New
	// Zealand.
	Model_JAPAC Model_HolidayRegion = 3
	// Europe, the Middle East and Africa.
	Model_EMEA Model_HolidayRegion = 4
	// Latin America and the Caribbean.
	Model_LAC Model_HolidayRegion = 5
	// United Arab Emirates
	Model_AE Model_HolidayRegion = 6
	// Argentina
	Model_AR Model_HolidayRegion = 7
	// Austria
	Model_AT Model_HolidayRegion = 8
	// Australia
	Model_AU Model_HolidayRegion = 9
	// Belgium
	Model_BE Model_HolidayRegion = 10
	// Brazil
	Model_BR Model_HolidayRegion = 11
	// Canada
	Model_CA Model_HolidayRegion = 12
	// Switzerland
	Model_CH Model_HolidayRegion = 13
	// Chile
	Model_CL Model_HolidayRegion = 14
	// China
	Model_CN Model_HolidayRegion = 15
	// Colombia
	Model_CO Model_HolidayRegion = 16
	// Czechoslovakia
	Model_CS Model_HolidayRegion = 17
	// Czech Republic
	Model_CZ Model_HolidayRegion = 18
	// Germany
	Model_DE Model_HolidayRegion = 19
	// Denmark
	Model_DK Model_HolidayRegion = 20
	// Algeria
	Model_DZ Model_HolidayRegion = 21
	// Ecuador
	Model_EC Model_HolidayRegion = 22
	// Estonia
	Model_EE Model_HolidayRegion = 23
	// Egypt
	Model_EG Model_HolidayRegion = 24
	// Spain
	Model_ES Model_HolidayRegion = 25
	// Finland
	Model_FI Model_HolidayRegion = 26
	// France
	Model_FR Model_HolidayRegion = 27
	// Great Britain (United Kingdom)
	Model_GB Model_HolidayRegion = 28
	// Greece
	Model_GR Model_HolidayRegion = 29
	// Hong Kong
	Model_HK Model_HolidayRegion = 30
	// Hungary
	Model_HU Model_HolidayRegion = 31
	// Indonesia
	Model_ID Model_HolidayRegion = 32
	// Ireland
	Model_IE Model_HolidayRegion = 33
	// Israel
	Model_IL Model_HolidayRegion = 34
	// India
	Model_IN Model_HolidayRegion = 35
	// Iran
	Model_IR Model_HolidayRegion = 36
	// Italy
	Model_IT Model_HolidayRegion = 37
	// Japan
	Model_JP Model_HolidayRegion = 38
	// Korea (South)
	Model_KR Model_HolidayRegion = 39
	// Latvia
	Model_LV Model_HolidayRegion = 40
	// Morocco
	Model_MA Model_HolidayRegion = 41
	// Mexico
	Model_MX Model_HolidayRegion = 42
	// Malaysia
	Model_MY Model_HolidayRegion = 43
	// Nigeria
	Model_NG Model_HolidayRegion = 44
	// Netherlands
	Model_NL Model_HolidayRegion = 45
	// Norway
	Model_NO Model_HolidayRegion = 46
	// New Zealand
	Model_NZ Model_HolidayRegion = 47
	// Peru
	Model_PE Model_HolidayRegion = 48
	// Philippines
	Model_PH Model_HolidayRegion = 49
	// Pakistan
	Model_PK Model_HolidayRegion = 50
	// Poland
	Model_PL Model_HolidayRegion = 51
	// Portugal
	Model_PT Model_HolidayRegion = 52
	// Romania
	Model_RO Model_HolidayRegion = 53
	// Serbia
	Model_RS Model_HolidayRegion = 54
	// Russian Federation
	Model_RU Model_HolidayRegion = 55
	// Saudi Arabia
	Model_SA Model_HolidayRegion = 56
	// Sweden
	Model_SE Model_HolidayRegion = 57
	// Singapore
	Model_SG Model_HolidayRegion = 58
	// Slovenia
	Model_SI Model_HolidayRegion = 59
	// Slovakia
	Model_SK Model_HolidayRegion = 60
	// Thailand
	Model_TH Model_HolidayRegion = 61
	// Turkey
	Model_TR Model_HolidayRegion = 62
	// Taiwan
	Model_TW Model_HolidayRegion = 63
	// Ukraine
	Model_UA Model_HolidayRegion = 64
	// United States
	Model_US Model_HolidayRegion = 65
	// Venezuela
	Model_VE Model_HolidayRegion = 66
	// Viet Nam
	Model_VN Model_HolidayRegion = 67
	// South Africa
	Model_ZA Model_HolidayRegion = 68
)

type Model_KmeansEnums struct {
	// contains filtered or unexported fields
}

type Model_KmeansEnums_KmeansInitializationMethod int32

const (
	// Unspecified initialization method.
	Model_KmeansEnums_KMEANS_INITIALIZATION_METHOD_UNSPECIFIED Model_KmeansEnums_KmeansInitializationMethod = 0
	// Initializes the centroids randomly.
	Model_KmeansEnums_RANDOM Model_KmeansEnums_KmeansInitializationMethod = 1
	// Initializes the centroids using data specified in
	// kmeans_initialization_column.
	Model_KmeansEnums_CUSTOM Model_KmeansEnums_KmeansInitializationMethod = 2
	// Initializes with kmeans++.
	Model_KmeansEnums_KMEANS_PLUS_PLUS Model_KmeansEnums_KmeansInitializationMethod = 3
)

type Model_LearnRateStrategy int32

const (
	Model_LEARN_RATE_STRATEGY_UNSPECIFIED Model_LearnRateStrategy = 0
	// Use line search to determine learning rate.
	Model_LINE_SEARCH Model_LearnRateStrategy = 1
	// Use a constant learning rate.
	Model_CONSTANT Model_LearnRateStrategy = 2
)

type Model_LossType int32

const (
	Model_LOSS_TYPE_UNSPECIFIED Model_LossType = 0
	// Mean squared loss, used for linear regression.
	Model_MEAN_SQUARED_LOSS Model_LossType = 1
	// Mean log loss, used for logistic regression.
	Model_MEAN_LOG_LOSS Model_LossType = 2
)

type Model_ModelType int32

const (
	Model_MODEL_TYPE_UNSPECIFIED Model_ModelType = 0
	// Linear regression model.
	Model_LINEAR_REGRESSION Model_ModelType = 1
	// Logistic regression based classification model.
	Model_LOGISTIC_REGRESSION Model_ModelType = 2
	// K-means clustering model.
	Model_KMEANS Model_ModelType = 3
	// Matrix factorization model.
	Model_MATRIX_FACTORIZATION Model_ModelType = 4
	// DNN classifier model.
	Model_DNN_CLASSIFIER Model_ModelType = 5
	// An imported TensorFlow model.
	Model_TENSORFLOW Model_ModelType = 6
	// DNN regressor model.
	Model_DNN_REGRESSOR Model_ModelType = 7
	// Boosted tree regressor model.
	Model_BOOSTED_TREE_REGRESSOR Model_ModelType = 9
	// Boosted tree classifier model.
	Model_BOOSTED_TREE_CLASSIFIER Model_ModelType = 10
	// ARIMA model.
	Model_ARIMA Model_ModelType = 11
	// [Beta] AutoML Tables regression model.
	Model_AUTOML_REGRESSOR Model_ModelType = 12
	// [Beta] AutoML Tables classification model.
	Model_AUTOML_CLASSIFIER Model_ModelType = 13
	// New name for the ARIMA model.
	Model_ARIMA_PLUS Model_ModelType = 19
)

type Model_MultiClassClassificationMetrics struct {

	// Aggregate classification metrics.
	AggregateClassificationMetrics *Model_AggregateClassificationMetrics `protobuf:"bytes,1,opt,name=aggregate_classification_metrics,json=aggregateClassificationMetrics,proto3" json:"aggregate_classification_metrics,omitempty"`
	// Confusion matrix at different thresholds.
	ConfusionMatrixList []*Model_MultiClassClassificationMetrics_ConfusionMatrix `protobuf:"bytes,2,rep,name=confusion_matrix_list,json=confusionMatrixList,proto3" json:"confusion_matrix_list,omitempty"`
	// contains filtered or unexported fields
}

type Model_MultiClassClassificationMetrics_ConfusionMatrix struct {

	// Confidence threshold used when computing the entries of the
	// confusion matrix.
	ConfidenceThreshold *wrapperspb.DoubleValue `protobuf:"bytes,1,opt,name=confidence_threshold,json=confidenceThreshold,proto3" json:"confidence_threshold,omitempty"`
	// One row per actual label.
	Rows []*Model_MultiClassClassificationMetrics_ConfusionMatrix_Row `protobuf:"bytes,2,rep,name=rows,proto3" json:"rows,omitempty"`
	// contains filtered or unexported fields
}

type Model_MultiClassClassificationMetrics_ConfusionMatrix_Entry struct {

	// The predicted label. For confidence_threshold > 0, we will
	// also add an entry indicating the number of items under the
	// confidence threshold.
	PredictedLabel string `protobuf:"bytes,1,opt,name=predicted_label,json=predictedLabel,proto3" json:"predicted_label,omitempty"`
	// Number of items being predicted as this label.
	ItemCount *wrapperspb.Int64Value `protobuf:"bytes,2,opt,name=item_count,json=itemCount,proto3" json:"item_count,omitempty"`
	// contains filtered or unexported fields
}

type Model_MultiClassClassificationMetrics_ConfusionMatrix_Row struct {

	// The original label of this row.
	ActualLabel string `protobuf:"bytes,1,opt,name=actual_label,json=actualLabel,proto3" json:"actual_label,omitempty"`
	// Info describing predicted label distribution.
	Entries []*Model_MultiClassClassificationMetrics_ConfusionMatrix_Entry `protobuf:"bytes,2,rep,name=entries,proto3" json:"entries,omitempty"`
	// contains filtered or unexported fields
}

type Model_OptimizationStrategy int32

const (
	Model_OPTIMIZATION_STRATEGY_UNSPECIFIED Model_OptimizationStrategy = 0
	// Uses an iterative batch gradient descent algorithm.
	Model_BATCH_GRADIENT_DESCENT Model_OptimizationStrategy = 1
	// Uses a normal equation to solve linear regression problem.
	Model_NORMAL_EQUATION Model_OptimizationStrategy = 2
)

type Model_RankingMetrics struct {

	// Calculates a precision per user for all the items by ranking them and
	// then averages all the precisions across all the users.
	MeanAveragePrecision *wrapperspb.DoubleValue `protobuf:"bytes,1,opt,name=mean_average_precision,json=meanAveragePrecision,proto3" json:"mean_average_precision,omitempty"`
	// Similar to the mean squared error computed in regression and explicit
	// recommendation models except instead of computing the rating directly,
	// the output from evaluate is computed against a preference which is 1 or 0
	// depending on if the rating exists or not.
	MeanSquaredError *wrapperspb.DoubleValue `protobuf:"bytes,2,opt,name=mean_squared_error,json=meanSquaredError,proto3" json:"mean_squared_error,omitempty"`
	// A metric to determine the goodness of a ranking calculated from the
	// predicted confidence by comparing it to an ideal rank measured by the
	// original ratings.
	NormalizedDiscountedCumulativeGain *wrapperspb.DoubleValue `protobuf:"bytes,3,opt,name=normalized_discounted_cumulative_gain,json=normalizedDiscountedCumulativeGain,proto3" json:"normalized_discounted_cumulative_gain,omitempty"`
	// Determines the goodness of a ranking by computing the percentile rank
	// from the predicted confidence and dividing it by the original rank.
	AverageRank *wrapperspb.DoubleValue `protobuf:"bytes,4,opt,name=average_rank,json=averageRank,proto3" json:"average_rank,omitempty"`
	// contains filtered or unexported fields
}

type Model_RegressionMetrics struct {

	// Mean absolute error.
	MeanAbsoluteError *wrapperspb.DoubleValue `protobuf:"bytes,1,opt,name=mean_absolute_error,json=meanAbsoluteError,proto3" json:"mean_absolute_error,omitempty"`
	// Mean squared error.
	MeanSquaredError *wrapperspb.DoubleValue `protobuf:"bytes,2,opt,name=mean_squared_error,json=meanSquaredError,proto3" json:"mean_squared_error,omitempty"`
	// Mean squared log error.
	MeanSquaredLogError *wrapperspb.DoubleValue `protobuf:"bytes,3,opt,name=mean_squared_log_error,json=meanSquaredLogError,proto3" json:"mean_squared_log_error,omitempty"`
	// Median absolute error.
	MedianAbsoluteError *wrapperspb.DoubleValue `protobuf:"bytes,4,opt,name=median_absolute_error,json=medianAbsoluteError,proto3" json:"median_absolute_error,omitempty"`
	// R^2 score. This corresponds to r2_score in ML.EVALUATE.
	RSquared *wrapperspb.DoubleValue `protobuf:"bytes,5,opt,name=r_squared,json=rSquared,proto3" json:"r_squared,omitempty"`
	// contains filtered or unexported fields
}

type Model_SeasonalPeriod struct {
	// contains filtered or unexported fields
}

type Model_SeasonalPeriod_SeasonalPeriodType int32

const (
	Model_SeasonalPeriod_SEASONAL_PERIOD_TYPE_UNSPECIFIED Model_SeasonalPeriod_SeasonalPeriodType = 0
	// No seasonality
	Model_SeasonalPeriod_NO_SEASONALITY Model_SeasonalPeriod_SeasonalPeriodType = 1
	// Daily period, 24 hours.
	Model_SeasonalPeriod_DAILY Model_SeasonalPeriod_SeasonalPeriodType = 2
	// Weekly period, 7 days.
	Model_SeasonalPeriod_WEEKLY Model_SeasonalPeriod_SeasonalPeriodType = 3
	// Monthly period, 30 days or irregular.
	Model_SeasonalPeriod_MONTHLY Model_SeasonalPeriod_SeasonalPeriodType = 4
	// Quarterly period, 90 days or irregular.
	Model_SeasonalPeriod_QUARTERLY Model_SeasonalPeriod_SeasonalPeriodType = 5
	// Yearly period, 365 days or irregular.
	Model_SeasonalPeriod_YEARLY Model_SeasonalPeriod_SeasonalPeriodType = 6
)

type Model_TrainingRun struct {

	// Options that were used for this training run, includes
	// user specified and default options that were used.
	TrainingOptions *Model_TrainingRun_TrainingOptions `protobuf:"bytes,1,opt,name=training_options,json=trainingOptions,proto3" json:"training_options,omitempty"`
	// The start time of this training run.
	StartTime *timestamppb.Timestamp `protobuf:"bytes,8,opt,name=start_time,json=startTime,proto3" json:"start_time,omitempty"`
	// Output of each iteration run, results.size() <= max_iterations.
	Results []*Model_TrainingRun_IterationResult `protobuf:"bytes,6,rep,name=results,proto3" json:"results,omitempty"`
	// The evaluation metrics over training/eval data that were computed at the
	// end of training.
	EvaluationMetrics *Model_EvaluationMetrics `protobuf:"bytes,7,opt,name=evaluation_metrics,json=evaluationMetrics,proto3" json:"evaluation_metrics,omitempty"`
	// Data split result of the training run. Only set when the input data is
	// actually split.
	DataSplitResult *Model_DataSplitResult `protobuf:"bytes,9,opt,name=data_split_result,json=dataSplitResult,proto3" json:"data_split_result,omitempty"`
	// Global explanations for important features of the model. For multi-class
	// models, there is one entry for each label class. For other models, there
	// is only one entry in the list.
	GlobalExplanations []*Model_GlobalExplanation `protobuf:"bytes,10,rep,name=global_explanations,json=globalExplanations,proto3" json:"global_explanations,omitempty"`
	// contains filtered or unexported fields
}

type Model_TrainingRun_IterationResult struct {

	// Index of the iteration, 0 based.
	Index *wrapperspb.Int32Value `protobuf:"bytes,1,opt,name=index,proto3" json:"index,omitempty"`
	// Time taken to run the iteration in milliseconds.
	DurationMs *wrapperspb.Int64Value `protobuf:"bytes,4,opt,name=duration_ms,json=durationMs,proto3" json:"duration_ms,omitempty"`
	// Loss computed on the training data at the end of iteration.
	TrainingLoss *wrapperspb.DoubleValue `protobuf:"bytes,5,opt,name=training_loss,json=trainingLoss,proto3" json:"training_loss,omitempty"`
	// Loss computed on the eval data at the end of iteration.
	EvalLoss *wrapperspb.DoubleValue `protobuf:"bytes,6,opt,name=eval_loss,json=evalLoss,proto3" json:"eval_loss,omitempty"`
	// Learn rate used for this iteration.
	LearnRate float64 `protobuf:"fixed64,7,opt,name=learn_rate,json=learnRate,proto3" json:"learn_rate,omitempty"`
	// Information about top clusters for clustering models.
	ClusterInfos []*Model_TrainingRun_IterationResult_ClusterInfo `protobuf:"bytes,8,rep,name=cluster_infos,json=clusterInfos,proto3" json:"cluster_infos,omitempty"`
	ArimaResult  *Model_TrainingRun_IterationResult_ArimaResult   `protobuf:"bytes,9,opt,name=arima_result,json=arimaResult,proto3" json:"arima_result,omitempty"`
	// contains filtered or unexported fields
}

type Model_TrainingRun_IterationResult_ArimaResult struct {

	// This message is repeated because there are multiple arima models
	// fitted in auto-arima. For non-auto-arima model, its size is one.
	ArimaModelInfo []*Model_TrainingRun_IterationResult_ArimaResult_ArimaModelInfo `protobuf:"bytes,1,rep,name=arima_model_info,json=arimaModelInfo,proto3" json:"arima_model_info,omitempty"`
	// Seasonal periods. Repeated because multiple periods are supported for
	// one time series.
	SeasonalPeriods []Model_SeasonalPeriod_SeasonalPeriodType `protobuf:"varint,2,rep,packed,name=seasonal_periods,json=seasonalPeriods,proto3,enum=google.cloud.bigquery.v2.Model_SeasonalPeriod_SeasonalPeriodType" json:"seasonal_periods,omitempty"`
	// contains filtered or unexported fields
}

type Model_TrainingRun_IterationResult_ArimaResult_ArimaCoefficients struct {

	// Auto-regressive coefficients, an array of double.
	AutoRegressiveCoefficients []float64 `protobuf:"fixed64,1,rep,packed,name=auto_regressive_coefficients,json=autoRegressiveCoefficients,proto3" json:"auto_regressive_coefficients,omitempty"`
	// Moving-average coefficients, an array of double.
	MovingAverageCoefficients []float64 `protobuf:"fixed64,2,rep,packed,name=moving_average_coefficients,json=movingAverageCoefficients,proto3" json:"moving_average_coefficients,omitempty"`
	// Intercept coefficient, just a double not an array.
	InterceptCoefficient float64 `protobuf:"fixed64,3,opt,name=intercept_coefficient,json=interceptCoefficient,proto3" json:"intercept_coefficient,omitempty"`
	// contains filtered or unexported fields
}

type Model_TrainingRun_IterationResult_ArimaResult_ArimaModelInfo struct {

	// Non-seasonal order.
	NonSeasonalOrder *Model_ArimaOrder `protobuf:"bytes,1,opt,name=non_seasonal_order,json=nonSeasonalOrder,proto3" json:"non_seasonal_order,omitempty"`
	// Arima coefficients.
	ArimaCoefficients *Model_TrainingRun_IterationResult_ArimaResult_ArimaCoefficients `protobuf:"bytes,2,opt,name=arima_coefficients,json=arimaCoefficients,proto3" json:"arima_coefficients,omitempty"`
	// Arima fitting metrics.
	ArimaFittingMetrics *Model_ArimaFittingMetrics `protobuf:"bytes,3,opt,name=arima_fitting_metrics,json=arimaFittingMetrics,proto3" json:"arima_fitting_metrics,omitempty"`
	// Whether Arima model fitted with drift or not. It is always false
	// when d is not 1.
	HasDrift bool `protobuf:"varint,4,opt,name=has_drift,json=hasDrift,proto3" json:"has_drift,omitempty"`
	// The time_series_id value for this time series. It will be one of
	// the unique values from the time_series_id_column specified during
	// ARIMA model training. Only present when time_series_id_column
	// training option was used.
	TimeSeriesId string `protobuf:"bytes,5,opt,name=time_series_id,json=timeSeriesId,proto3" json:"time_series_id,omitempty"`
	// The tuple of time_series_ids identifying this time series. It will
	// be one of the unique tuples of values present in the
	// time_series_id_columns specified during ARIMA model training. Only
	// present when time_series_id_columns training option was used and
	// the order of values here are same as the order of
	// time_series_id_columns.
	TimeSeriesIds []string `protobuf:"bytes,10,rep,name=time_series_ids,json=timeSeriesIds,proto3" json:"time_series_ids,omitempty"`
	// Seasonal periods. Repeated because multiple periods are supported
	// for one time series.
	SeasonalPeriods []Model_SeasonalPeriod_SeasonalPeriodType `protobuf:"varint,6,rep,packed,name=seasonal_periods,json=seasonalPeriods,proto3,enum=google.cloud.bigquery.v2.Model_SeasonalPeriod_SeasonalPeriodType" json:"seasonal_periods,omitempty"`
	// If true, holiday_effect is a part of time series decomposition
	// result.
	HasHolidayEffect *wrapperspb.BoolValue `protobuf:"bytes,7,opt,name=has_holiday_effect,json=hasHolidayEffect,proto3" json:"has_holiday_effect,omitempty"`
	// If true, spikes_and_dips is a part of time series decomposition
	// result.
	HasSpikesAndDips *wrapperspb.BoolValue `protobuf:"bytes,8,opt,name=has_spikes_and_dips,json=hasSpikesAndDips,proto3" json:"has_spikes_and_dips,omitempty"`
	// If true, step_changes is a part of time series decomposition
	// result.
	HasStepChanges *wrapperspb.BoolValue `protobuf:"bytes,9,opt,name=has_step_changes,json=hasStepChanges,proto3" json:"has_step_changes,omitempty"`
	// contains filtered or unexported fields
}

type Model_TrainingRun_IterationResult_ClusterInfo struct {

	// Centroid id.
	CentroidId int64 `protobuf:"varint,1,opt,name=centroid_id,json=centroidId,proto3" json:"centroid_id,omitempty"`
	// Cluster radius, the average distance from centroid
	// to each point assigned to the cluster.
	ClusterRadius *wrapperspb.DoubleValue `protobuf:"bytes,2,opt,name=cluster_radius,json=clusterRadius,proto3" json:"cluster_radius,omitempty"`
	// Cluster size, the total number of points assigned to the cluster.
	ClusterSize *wrapperspb.Int64Value `protobuf:"bytes,3,opt,name=cluster_size,json=clusterSize,proto3" json:"cluster_size,omitempty"`
	// contains filtered or unexported fields
}

type Model_TrainingRun_TrainingOptions struct {

	// The maximum number of iterations in training. Used only for iterative
	// training algorithms.
	MaxIterations int64 `protobuf:"varint,1,opt,name=max_iterations,json=maxIterations,proto3" json:"max_iterations,omitempty"`
	// Type of loss function used during training run.
	LossType Model_LossType `protobuf:"varint,2,opt,name=loss_type,json=lossType,proto3,enum=google.cloud.bigquery.v2.Model_LossType" json:"loss_type,omitempty"`
	// Learning rate in training. Used only for iterative training algorithms.
	LearnRate float64 `protobuf:"fixed64,3,opt,name=learn_rate,json=learnRate,proto3" json:"learn_rate,omitempty"`
	// L1 regularization coefficient.
	L1Regularization *wrapperspb.DoubleValue `protobuf:"bytes,4,opt,name=l1_regularization,json=l1Regularization,proto3" json:"l1_regularization,omitempty"`
	// L2 regularization coefficient.
	L2Regularization *wrapperspb.DoubleValue `protobuf:"bytes,5,opt,name=l2_regularization,json=l2Regularization,proto3" json:"l2_regularization,omitempty"`
	// When early_stop is true, stops training when accuracy improvement is
	// less than 'min_relative_progress'. Used only for iterative training
	// algorithms.
	MinRelativeProgress *wrapperspb.DoubleValue `protobuf:"bytes,6,opt,name=min_relative_progress,json=minRelativeProgress,proto3" json:"min_relative_progress,omitempty"`
	// Whether to train a model from the last checkpoint.
	WarmStart *wrapperspb.BoolValue `protobuf:"bytes,7,opt,name=warm_start,json=warmStart,proto3" json:"warm_start,omitempty"`
	// Whether to stop early when the loss doesn't improve significantly
	// any more (compared to min_relative_progress). Used only for iterative
	// training algorithms.
	EarlyStop *wrapperspb.BoolValue `protobuf:"bytes,8,opt,name=early_stop,json=earlyStop,proto3" json:"early_stop,omitempty"`
	// Name of input label columns in training data.
	InputLabelColumns []string `protobuf:"bytes,9,rep,name=input_label_columns,json=inputLabelColumns,proto3" json:"input_label_columns,omitempty"`
	// The data split type for training and evaluation, e.g. RANDOM.
	DataSplitMethod Model_DataSplitMethod `protobuf:"varint,10,opt,name=data_split_method,json=dataSplitMethod,proto3,enum=google.cloud.bigquery.v2.Model_DataSplitMethod" json:"data_split_method,omitempty"`
	// The fraction of evaluation data over the whole input data. The rest
	// of data will be used as training data. The format should be double.
	// Accurate to two decimal places.
	// Default value is 0.2.
	DataSplitEvalFraction float64 `protobuf:"fixed64,11,opt,name=data_split_eval_fraction,json=dataSplitEvalFraction,proto3" json:"data_split_eval_fraction,omitempty"`
	// The column to split data with. This column won't be used as a
	// feature.
	// 1. When data_split_method is CUSTOM, the corresponding column should
	// be boolean. The rows with true value tag are eval data, and the false
	// are training data.
	// 2. When data_split_method is SEQ, the first DATA_SPLIT_EVAL_FRACTION
	// rows (from smallest to largest) in the corresponding column are used
	// as training data, and the rest are eval data. It respects the order
	// in Orderable data types:
	// https://cloud.google.com/bigquery/docs/reference/standard-sql/data-types#data-type-properties
	DataSplitColumn string `protobuf:"bytes,12,opt,name=data_split_column,json=dataSplitColumn,proto3" json:"data_split_column,omitempty"`
	// The strategy to determine learn rate for the current iteration.
	LearnRateStrategy Model_LearnRateStrategy `protobuf:"varint,13,opt,name=learn_rate_strategy,json=learnRateStrategy,proto3,enum=google.cloud.bigquery.v2.Model_LearnRateStrategy" json:"learn_rate_strategy,omitempty"`
	// Specifies the initial learning rate for the line search learn rate
	// strategy.
	InitialLearnRate float64 `protobuf:"fixed64,16,opt,name=initial_learn_rate,json=initialLearnRate,proto3" json:"initial_learn_rate,omitempty"`
	// Weights associated with each label class, for rebalancing the
	// training data. Only applicable for classification models.
	LabelClassWeights map[string]float64 `protobuf:"bytes,17,rep,name=label_class_weights,json=labelClassWeights,proto3" json:"label_class_weights,omitempty" protobuf_key:"bytes,1,opt,name=key,proto3" protobuf_val:"fixed64,2,opt,name=value,proto3"`
	// User column specified for matrix factorization models.
	UserColumn string `protobuf:"bytes,18,opt,name=user_column,json=userColumn,proto3" json:"user_column,omitempty"`
	// Item column specified for matrix factorization models.
	ItemColumn string `protobuf:"bytes,19,opt,name=item_column,json=itemColumn,proto3" json:"item_column,omitempty"`
	// Distance type for clustering models.
	DistanceType Model_DistanceType `protobuf:"varint,20,opt,name=distance_type,json=distanceType,proto3,enum=google.cloud.bigquery.v2.Model_DistanceType" json:"distance_type,omitempty"`
	// Number of clusters for clustering models.
	NumClusters int64 `protobuf:"varint,21,opt,name=num_clusters,json=numClusters,proto3" json:"num_clusters,omitempty"`
	// Google Cloud Storage URI from which the model was imported. Only
	// applicable for imported models.
	ModelUri string `protobuf:"bytes,22,opt,name=model_uri,json=modelUri,proto3" json:"model_uri,omitempty"`
	// Optimization strategy for training linear regression models.
	OptimizationStrategy Model_OptimizationStrategy `protobuf:"varint,23,opt,name=optimization_strategy,json=optimizationStrategy,proto3,enum=google.cloud.bigquery.v2.Model_OptimizationStrategy" json:"optimization_strategy,omitempty"`
	// Hidden units for dnn models.
	HiddenUnits []int64 `protobuf:"varint,24,rep,packed,name=hidden_units,json=hiddenUnits,proto3" json:"hidden_units,omitempty"`
	// Batch size for dnn models.
	BatchSize int64 `protobuf:"varint,25,opt,name=batch_size,json=batchSize,proto3" json:"batch_size,omitempty"`
	// Dropout probability for dnn models.
	Dropout *wrapperspb.DoubleValue `protobuf:"bytes,26,opt,name=dropout,proto3" json:"dropout,omitempty"`
	// Maximum depth of a tree for boosted tree models.
	MaxTreeDepth int64 `protobuf:"varint,27,opt,name=max_tree_depth,json=maxTreeDepth,proto3" json:"max_tree_depth,omitempty"`
	// Subsample fraction of the training data to grow tree to prevent
	// overfitting for boosted tree models.
	Subsample float64 `protobuf:"fixed64,28,opt,name=subsample,proto3" json:"subsample,omitempty"`
	// Minimum split loss for boosted tree models.
	MinSplitLoss *wrapperspb.DoubleValue `protobuf:"bytes,29,opt,name=min_split_loss,json=minSplitLoss,proto3" json:"min_split_loss,omitempty"`
	// Num factors specified for matrix factorization models.
	NumFactors int64 `protobuf:"varint,30,opt,name=num_factors,json=numFactors,proto3" json:"num_factors,omitempty"`
	// Feedback type that specifies which algorithm to run for matrix
	// factorization.
	FeedbackType Model_FeedbackType `protobuf:"varint,31,opt,name=feedback_type,json=feedbackType,proto3,enum=google.cloud.bigquery.v2.Model_FeedbackType" json:"feedback_type,omitempty"`
	// Hyperparameter for matrix factoration when implicit feedback type is
	// specified.
	WalsAlpha *wrapperspb.DoubleValue `protobuf:"bytes,32,opt,name=wals_alpha,json=walsAlpha,proto3" json:"wals_alpha,omitempty"`
	// The method used to initialize the centroids for kmeans algorithm.
	KmeansInitializationMethod Model_KmeansEnums_KmeansInitializationMethod `protobuf:"varint,33,opt,name=kmeans_initialization_method,json=kmeansInitializationMethod,proto3,enum=google.cloud.bigquery.v2.Model_KmeansEnums_KmeansInitializationMethod" json:"kmeans_initialization_method,omitempty"`
	// The column used to provide the initial centroids for kmeans algorithm
	// when kmeans_initialization_method is CUSTOM.
	KmeansInitializationColumn string `protobuf:"bytes,34,opt,name=kmeans_initialization_column,json=kmeansInitializationColumn,proto3" json:"kmeans_initialization_column,omitempty"`
	// Column to be designated as time series timestamp for ARIMA model.
	TimeSeriesTimestampColumn string `protobuf:"bytes,35,opt,name=time_series_timestamp_column,json=timeSeriesTimestampColumn,proto3" json:"time_series_timestamp_column,omitempty"`
	// Column to be designated as time series data for ARIMA model.
	TimeSeriesDataColumn string `protobuf:"bytes,36,opt,name=time_series_data_column,json=timeSeriesDataColumn,proto3" json:"time_series_data_column,omitempty"`
	// Whether to enable auto ARIMA or not.
	AutoArima bool `protobuf:"varint,37,opt,name=auto_arima,json=autoArima,proto3" json:"auto_arima,omitempty"`
	// A specification of the non-seasonal part of the ARIMA model: the three
	// components (p, d, q) are the AR order, the degree of differencing, and
	// the MA order.
	NonSeasonalOrder *Model_ArimaOrder `protobuf:"bytes,38,opt,name=non_seasonal_order,json=nonSeasonalOrder,proto3" json:"non_seasonal_order,omitempty"`
	// The data frequency of a time series.
	DataFrequency Model_DataFrequency `protobuf:"varint,39,opt,name=data_frequency,json=dataFrequency,proto3,enum=google.cloud.bigquery.v2.Model_DataFrequency" json:"data_frequency,omitempty"`
	// Include drift when fitting an ARIMA model.
	IncludeDrift bool `protobuf:"varint,41,opt,name=include_drift,json=includeDrift,proto3" json:"include_drift,omitempty"`
	// The geographical region based on which the holidays are considered in
	// time series modeling. If a valid value is specified, then holiday
	// effects modeling is enabled.
	HolidayRegion Model_HolidayRegion `protobuf:"varint,42,opt,name=holiday_region,json=holidayRegion,proto3,enum=google.cloud.bigquery.v2.Model_HolidayRegion" json:"holiday_region,omitempty"`
	// The time series id column that was used during ARIMA model training.
	TimeSeriesIdColumn string `protobuf:"bytes,43,opt,name=time_series_id_column,json=timeSeriesIdColumn,proto3" json:"time_series_id_column,omitempty"`
	// The time series id columns that were used during ARIMA model training.
	TimeSeriesIdColumns []string `protobuf:"bytes,51,rep,name=time_series_id_columns,json=timeSeriesIdColumns,proto3" json:"time_series_id_columns,omitempty"`
	// The number of periods ahead that need to be forecasted.
	Horizon int64 `protobuf:"varint,44,opt,name=horizon,proto3" json:"horizon,omitempty"`
	// Whether to preserve the input structs in output feature names.
	// Suppose there is a struct A with field b.
	// When false (default), the output feature name is A_b.
	// When true, the output feature name is A.b.
	PreserveInputStructs bool `protobuf:"varint,45,opt,name=preserve_input_structs,json=preserveInputStructs,proto3" json:"preserve_input_structs,omitempty"`
	// The max value of non-seasonal p and q.
	AutoArimaMaxOrder int64 `protobuf:"varint,46,opt,name=auto_arima_max_order,json=autoArimaMaxOrder,proto3" json:"auto_arima_max_order,omitempty"`
	// If true, perform decompose time series and save the results.
	DecomposeTimeSeries *wrapperspb.BoolValue `protobuf:"bytes,50,opt,name=decompose_time_series,json=decomposeTimeSeries,proto3" json:"decompose_time_series,omitempty"`
	// If true, clean spikes and dips in the input time series.
	CleanSpikesAndDips *wrapperspb.BoolValue `protobuf:"bytes,52,opt,name=clean_spikes_and_dips,json=cleanSpikesAndDips,proto3" json:"clean_spikes_and_dips,omitempty"`
	// If true, detect step changes and make data adjustment in the input time
	// series.
	AdjustStepChanges *wrapperspb.BoolValue `protobuf:"bytes,53,opt,name=adjust_step_changes,json=adjustStepChanges,proto3" json:"adjust_step_changes,omitempty"`
	// contains filtered or unexported fields
}

type PatchModelRequest struct {

	// Required. Project ID of the model to patch.
	ProjectId string `protobuf:"bytes,1,opt,name=project_id,json=projectId,proto3" json:"project_id,omitempty"`
	// Required. Dataset ID of the model to patch.
	DatasetId string `protobuf:"bytes,2,opt,name=dataset_id,json=datasetId,proto3" json:"dataset_id,omitempty"`
	// Required. Model ID of the model to patch.
	ModelId string `protobuf:"bytes,3,opt,name=model_id,json=modelId,proto3" json:"model_id,omitempty"`
	// Required. Patched model.
	// Follows RFC5789 patch semantics. Missing fields are not updated.
	// To clear a field, explicitly set to default value.
	Model *Model `protobuf:"bytes,4,opt,name=model,proto3" json:"model,omitempty"`
	// contains filtered or unexported fields
}

type StandardSqlDataType struct {

	// Required. The top level type of this field.
	// Can be any standard SQL data type (e.g., "INT64", "DATE", "ARRAY").
	TypeKind StandardSqlDataType_TypeKind `protobuf:"varint,1,opt,name=type_kind,json=typeKind,proto3,enum=google.cloud.bigquery.v2.StandardSqlDataType_TypeKind" json:"type_kind,omitempty"`
	// Types that are assignable to SubType:
	//	*StandardSqlDataType_ArrayElementType
	//	*StandardSqlDataType_StructType
	SubType isStandardSqlDataType_SubType `protobuf_oneof:"sub_type"`
	// contains filtered or unexported fields
}

type StandardSqlDataType_ArrayElementType struct {
	// The type of the array's elements, if type_kind = "ARRAY".
	ArrayElementType *StandardSqlDataType `protobuf:"bytes,2,opt,name=array_element_type,json=arrayElementType,proto3,oneof"`
}

type StandardSqlDataType_StructType struct {
	// The fields of this struct, in order, if type_kind = "STRUCT".
	StructType *StandardSqlStructType `protobuf:"bytes,3,opt,name=struct_type,json=structType,proto3,oneof"`
}

type StandardSqlDataType_TypeKind int32

const (
	// Invalid type.
	StandardSqlDataType_TYPE_KIND_UNSPECIFIED StandardSqlDataType_TypeKind = 0
	// Encoded as a string in decimal format.
	StandardSqlDataType_INT64 StandardSqlDataType_TypeKind = 2
	// Encoded as a boolean "false" or "true".
	StandardSqlDataType_BOOL StandardSqlDataType_TypeKind = 5
	// Encoded as a number, or string "NaN", "Infinity" or "-Infinity".
	StandardSqlDataType_FLOAT64 StandardSqlDataType_TypeKind = 7
	// Encoded as a string value.
	StandardSqlDataType_STRING StandardSqlDataType_TypeKind = 8
	// Encoded as a base64 string per RFC 4648, section 4.
	StandardSqlDataType_BYTES StandardSqlDataType_TypeKind = 9
	// Encoded as an RFC 3339 timestamp with mandatory "Z" time zone string:
	// 1985-04-12T23:20:50.52Z
	StandardSqlDataType_TIMESTAMP StandardSqlDataType_TypeKind = 19
	// Encoded as RFC 3339 full-date format string: 1985-04-12
	StandardSqlDataType_DATE StandardSqlDataType_TypeKind = 10
	// Encoded as RFC 3339 partial-time format string: 23:20:50.52
	StandardSqlDataType_TIME StandardSqlDataType_TypeKind = 20
	// Encoded as RFC 3339 full-date "T" partial-time: 1985-04-12T23:20:50.52
	StandardSqlDataType_DATETIME StandardSqlDataType_TypeKind = 21
	// Encoded as fully qualified 3 part: 0-5 15 2:30:45.6
	StandardSqlDataType_INTERVAL StandardSqlDataType_TypeKind = 26
	// Encoded as WKT
	StandardSqlDataType_GEOGRAPHY StandardSqlDataType_TypeKind = 22
	// Encoded as a decimal string.
	StandardSqlDataType_NUMERIC StandardSqlDataType_TypeKind = 23
	// Encoded as a decimal string.
	StandardSqlDataType_BIGNUMERIC StandardSqlDataType_TypeKind = 24
	// Encoded as a string.
	StandardSqlDataType_JSON StandardSqlDataType_TypeKind = 25
	// Encoded as a list with types matching Type.array_type.
	StandardSqlDataType_ARRAY StandardSqlDataType_TypeKind = 16
	// Encoded as a list with fields of type Type.struct_type[i]. List is used
	// because a JSON object cannot have duplicate field names.
	StandardSqlDataType_STRUCT StandardSqlDataType_TypeKind = 17
)

type StandardSqlField struct {

	// Optional. The name of this field. Can be absent for struct fields.
	Name string `protobuf:"bytes,1,opt,name=name,proto3" json:"name,omitempty"`
	// Optional. The type of this parameter. Absent if not explicitly
	// specified (e.g., CREATE FUNCTION statement can omit the return type;
	// in this case the output parameter does not have this "type" field).
	Type *StandardSqlDataType `protobuf:"bytes,2,opt,name=type,proto3" json:"type,omitempty"`
	// contains filtered or unexported fields
}

type StandardSqlStructType struct {
	Fields []*StandardSqlField `protobuf:"bytes,1,rep,name=fields,proto3" json:"fields,omitempty"`
	// contains filtered or unexported fields
}

type StandardSqlTableType struct {

	// The columns in this table type
	Columns []*StandardSqlField `protobuf:"bytes,1,rep,name=columns,proto3" json:"columns,omitempty"`
	// contains filtered or unexported fields
}

type TableReference struct {

	// Required. The ID of the project containing this table.
	ProjectId string `protobuf:"bytes,1,opt,name=project_id,json=projectId,proto3" json:"project_id,omitempty"`
	// Required. The ID of the dataset containing this table.
	DatasetId string `protobuf:"bytes,2,opt,name=dataset_id,json=datasetId,proto3" json:"dataset_id,omitempty"`
	// Required. The ID of the table. The ID must contain only
	// letters (a-z, A-Z), numbers (0-9), or underscores (_). The maximum
	// length is 1,024 characters.  Certain operations allow
	// suffixing of the table ID with a partition decorator, such as
	// `sample_table$20190123`.
	TableId string `protobuf:"bytes,3,opt,name=table_id,json=tableId,proto3" json:"table_id,omitempty"`
	// The alternative field that will be used when ESF is not able to translate
	// the received data to the project_id field.
	ProjectIdAlternative []string `protobuf:"bytes,4,rep,name=project_id_alternative,json=projectIdAlternative,proto3" json:"project_id_alternative,omitempty"`
	// The alternative field that will be used when ESF is not able to translate
	// the received data to the project_id field.
	DatasetIdAlternative []string `protobuf:"bytes,5,rep,name=dataset_id_alternative,json=datasetIdAlternative,proto3" json:"dataset_id_alternative,omitempty"`
	// The alternative field that will be used when ESF is not able to translate
	// the received data to the project_id field.
	TableIdAlternative []string `protobuf:"bytes,6,rep,name=table_id_alternative,json=tableIdAlternative,proto3" json:"table_id_alternative,omitempty"`
	// contains filtered or unexported fields
}

type UnimplementedModelServiceServer struct {
}