Google Cloud Artifact Types¶

Artifact types corresponding to Google Cloud Resources produced and consumed by GCPC components.

These artifact types can be used in your custom KFP SDK components similarly to other KFP SDK artifacts. If you wish to produce Google artifacts from your own components, it is recommended that you use Containerized Python Components. You should assign metadata to the Google artifacts according to the artifact’s schema (provided by each artifact’s .schema attribute).

Classes:

class google_cloud_pipeline_components.types.artifact_types.BQMLModel(name: str | None = None, uri: str | None = None, metadata: dict | None = None)[source]¶

Bases: Artifact

An artifact representing a Google Cloud BQML Model resource.

Methods:

create(name, project_id, dataset_id, model_id)

Create a BQMLModel artifact instance.

Attributes:

schema
schema_title
schema_version

classmethod create(name: str, project_id: str, dataset_id: str, model_id: str) → BQMLModel[source]¶

Create a BQMLModel artifact instance.

Parameters¶

name: str¶

The artifact name.

project_id: str¶

The ID of the project containing this model.

dataset_id: str¶

The ID of the dataset containing this model.

model_id: str¶

The ID of the model. For more details, see

https

//cloud.google.com/bigquery/docs/reference/rest/v2/models#ModelReference

schema = 'title: google.BQMLModel\ntype: object\nproperties:\n projectId:\n type: string\n datasetId:\n type: string\n modelId:\n type: string'¶

schema_title = 'google.BQMLModel'¶

schema_version = '0.0.1'¶

class google_cloud_pipeline_components.types.artifact_types.BQTable(name: str | None = None, uri: str | None = None, metadata: dict | None = None)[source]¶

Bases: Artifact

An artifact representing a Google Cloud BQ Table resource.

Methods:

create(name, project_id, dataset_id, table_id)

Create a BQTable artifact instance.

Attributes:

schema
schema_title
schema_version

classmethod create(name: str, project_id: str, dataset_id: str, table_id: str) → BQTable[source]¶

Create a BQTable artifact instance.

Parameters¶

name: str¶

The artifact name.

project_id: str¶

The ID of the project containing this table.

dataset_id: str¶

The ID of the dataset containing this table.

table_id: str¶

The ID of the table. For more details, see

https

//cloud.google.com/bigquery/docs/reference/rest/v2/TableReference

schema = 'title: google.BQTable\ntype: object\nproperties:\n projectId:\n type: string\n datasetId:\n type: string\n tableId:\n type: string\n expirationTime:\n type: string'¶

schema_title = 'google.BQTable'¶

schema_version = '0.0.1'¶

class google_cloud_pipeline_components.types.artifact_types.ClassificationMetrics(name: str | None = None, uri: str | None = None, metadata: dict | None = None)[source]¶

Bases: Artifact

An artifact representing evaluation classification metrics.

Methods:

create([name, recall, precision, f1_score, ...])

Create a ClassificationMetrics artifact instance.

Attributes:

schema
schema_title
schema_version

classmethod create(name: str = 'evaluation_metrics', recall: float | None = None, precision: float | None = None, f1_score: float | None = None, accuracy: float | None = None, au_prc: float | None = None, au_roc: float | None = None, log_loss: float | None = None) → ClassificationMetrics[source]¶

Create a ClassificationMetrics artifact instance.

Parameters¶

recall: float | None = None¶

Recall (True Positive Rate) for the given confidence threshold.

precision: float | None = None¶

Precision for the given confidence threshold.

f1_score: float | None = None¶

The harmonic mean of recall and precision.

accuracy: float | None = None¶

Accuracy is the fraction of predictions given the correct label.

au_prc: float | None = None¶

The Area Under Precision-Recall Curve metric.

au_roc: float | None = None¶

The Area Under Receiver Operating Characteristic curve metric.

log_loss: float | None = None¶

The Log Loss metric.

schema = 'title: google.ClassificationMetrics\ntype: object\nproperties:\n aggregationType:\n type: string\n enum:\n - AGGREGATION_TYPE_UNSPECIFIED\n - MACRO_AVERAGE\n - MICRO_AVERAGE\n aggregationThreshold:\n type: number\n format: float\n recall:\n type: number\n format: float\n precision:\n type: number\n format: float\n f1_score:\n type: number\n format: float\n accuracy:\n type: number\n format: float\n auPrc:\n type: number\n format: float\n auRoc:\n type: number\n format: float\n logLoss:\n type: number\n format: float\n confusionMatrix:\n type: object\n properties:\n rows:\n type: array\n items:\n type: array\n items:\n type: integer\n format: int64\n annotationSpecs:\n type: array\n items:\n type: object\n properties:\n id:\n type: string\n displayName:\n type: string\n confidenceMetrics:\n type: array\n items:\n type: object\n properties:\n confidenceThreshold:\n type: number\n format: float\n recall:\n type: number\n format: float\n precision:\n type: number\n format: float\n f1Score:\n type: number\n format: float\n maxPredictions:\n type: integer\n format: int32\n falsePositiveRate:\n type: number\n format: float\n accuracy:\n type: number\n format: float\n truePositiveCount:\n type: integer\n format: int64\n falsePositiveCount:\n type: integer\n format: int64\n falseNegativeCount:\n type: integer\n format: int64\n trueNegativeCount:\n type: integer\n format: int64\n recallAt1:\n type: number\n format: float\n precisionAt1:\n type: number\n format: float\n falsePositiveRateAt1:\n type: number\n format: float\n f1ScoreAt1:\n type: number\n format: float\n confusionMatrix:\n type: object\n properties:\n rows:\n type: array\n items:\n type: array\n items:\n type: integer\n format: int64\n annotationSpecs:\n type: array\n items:\n type: object\n properties:\n id:\n type: string\n displayName:\n type: string'¶

schema_title = 'google.ClassificationMetrics'¶

schema_version = '0.0.1'¶

class google_cloud_pipeline_components.types.artifact_types.ForecastingMetrics(name: str | None = None, uri: str | None = None, metadata: dict | None = None)[source]¶

Bases: Artifact

An artifact representing evaluation forecasting metrics.

Methods:

create([name, root_mean_squared_error, ...])

Create a ForecastingMetrics artifact instance.

Attributes:

schema
schema_title
schema_version

classmethod create(name: str = 'evaluation_metrics', root_mean_squared_error: float | None = None, mean_absolute_error: float | None = None, mean_absolute_percentage_error: float | None = None, r_squared: float | None = None, root_mean_squared_log_error: float | None = None, weighted_absolute_percentage_error: float | None = None, root_mean_squared_percentage_error: float | None = None, symmetric_mean_absolute_percentage_error: float | None = None) → ForecastingMetrics[source]¶

Create a ForecastingMetrics artifact instance.

Parameters¶

root_mean_squared_error: float | None = None¶

Root Mean Squared Error (RMSE).

mean_absolute_error: float | None = None¶

Mean Absolute Error (MAE).

mean_absolute_percentage_error: float | None = None¶

Mean absolute percentage error.

r_squared: float | None = None¶

Coefficient of determination as Pearson correlation coefficient.

root_mean_squared_log_error: float | None = None¶

Root mean squared log error.

weighted_absolute_percentage_error: float | None = None¶

Weighted Absolute Percentage Error. Does not use weights, this is just what the metric is called. Undefined if actual values sum to zero. Will be very large if actual values sum to a very small number.

root_mean_squared_percentage_error: float | None = None¶

Root Mean Square Percentage Error. Square root of MSPE. Undefined/imaginary when MSPE is negative.

symmetric_mean_absolute_percentage_error: float | None = None¶

Symmetric Mean Absolute Percentage Error.

schema = 'title: google.ForecastingMetrics\ntype: object\nproperties:\n rootMeanSquaredError:\n type: number\n format: float\n meanAbsoluteError:\n type: number\n format: float\n meanAbsolutePercentageError:\n type: number\n format: float\n rSquared:\n type: number\n format: float\n rootMeanSquaredLogError:\n type: number\n format: float\n weightedAbsolutePercentageError:\n type: number\n format: float\n rootMeanSquaredPercentageError:\n type: number\n format: float\n symmetricMeanAbsolutePercentageError:\n type: number\n format: float\n quantileMetrics:\n type: array\n items:\n type: object\n properties:\n quantile:\n type: number\n format: double\n scaledPinballLoss:\n type: number\n format: float\n observedQuantile:\n type: number\n format: double'¶

schema_title = 'google.ForecastingMetrics'¶

schema_version = '0.0.1'¶

class google_cloud_pipeline_components.types.artifact_types.RegressionMetrics(name: str | None = None, uri: str | None = None, metadata: dict | None = None)[source]¶

Bases: Artifact

An artifact representing evaluation regression metrics.

Methods:

create([name, root_mean_squared_error, ...])

Create a RegressionMetrics artifact instance.

Attributes:

schema
schema_title
schema_version

classmethod create(name: str = 'evaluation_metrics', root_mean_squared_error: float | None = None, mean_absolute_error: float | None = None, mean_absolute_percentage_error: float | None = None, r_squared: float | None = None, root_mean_squared_log_error: float | None = None) → RegressionMetrics[source]¶

Create a RegressionMetrics artifact instance.

Parameters¶

root_mean_squared_error: float | None = None¶

Root Mean Squared Error (RMSE).

mean_absolute_error: float | None = None¶

Mean Absolute Error (MAE).

mean_absolute_percentage_error: float | None = None¶

Mean absolute percentage error.

r_squared: float | None = None¶

Coefficient of determination as Pearson correlation coefficient.

root_mean_squared_log_error: float | None = None¶

Root mean squared log error.

schema = 'title: google.RegressionMetrics\ntype: object\nproperties:\n rootMeanSquaredError:\n type: number\n format: float\n meanAbsoluteError:\n type: number\n format: float\n meanAbsolutePercentageError:\n type: number\n format: float\n rSquared:\n type: number\n format: float\n rootMeanSquaredLogError:\n type: number\n format: float'¶

schema_title = 'google.RegressionMetrics'¶

schema_version = '0.0.1'¶

class google_cloud_pipeline_components.types.artifact_types.UnmanagedContainerModel(name: str | None = None, uri: str | None = None, metadata: dict | None = None)[source]¶

Bases: Artifact

An artifact representing a Vertex AI unmanaged container model.

Methods:

create(predict_schemata, container_spec)

Create a UnmanagedContainerModel artifact instance.

Attributes:

schema
schema_title
schema_version

classmethod create(predict_schemata: dict, container_spec: dict) → UnmanagedContainerModel[source]¶

Create a UnmanagedContainerModel artifact instance.

Parameters¶

predict_schemata: dict¶

Contains the schemata used in Model’s predictions and explanations via PredictionService.Predict, PredictionService.Explain and BatchPredictionJob. For more details, see

https

//cloud.google.com/vertex-ai/docs/reference/rest/v1/PredictSchemata

container_spec: dict¶

Specification of a container for serving predictions. Some fields in this message correspond to fields in the Kubernetes Container v1 core specification. For more details, see

https

//cloud.google.com/vertex-ai/docs/reference/rest/v1/ModelContainerSpec

schema = 'title: google.UnmanagedContainerModel\ntype: object\nproperties:\n predictSchemata:\n type: object\n properties:\n instanceSchemaUri:\n type: string\n parametersSchemaUri:\n type: string\n predictionSchemaUri:\n type: string\n containerSpec:\n type: object\n properties:\n imageUri:\n type: string\n command:\n type: array\n items:\n type: string\n args:\n type: array\n items:\n type: string\n env:\n type: array\n items:\n type: object\n properties:\n name:\n type: string\n value:\n type: string\n ports:\n type: array\n items:\n type: object\n properties:\n containerPort:\n type: integer\n predictRoute:\n type: string\n healthRoute:\n type: string'¶

schema_title = 'google.UnmanagedContainerModel'¶

schema_version = '0.0.1'¶

class google_cloud_pipeline_components.types.artifact_types.VertexBatchPredictionJob(name: str | None = None, uri: str | None = None, metadata: dict | None = None)[source]¶

Bases: Artifact

An artifact representing a Vertex AI BatchPredictionJob resource.

Methods:

create(name, uri, job_resource_name[, ...])

Create a VertexBatchPredictionJob artifact instance.

Attributes:

schema
schema_title
schema_version

classmethod create(name: str, uri: str, job_resource_name: str, bigquery_output_table: str | None = None, bigquery_output_dataset: str | None = None, gcs_output_directory: str | None = None) → VertexBatchPredictionJob[source]¶

Create a VertexBatchPredictionJob artifact instance.

Parameters¶

name: str¶

The artifact name.

uri: str¶

the Vertex Batch Prediction resource uri, in a form of

https

//{service-endpoint}/v1/projects/{project}/locations/{location}/batchPredictionJobs/{batchPredictionJob}, where {service-endpoint} is one of the supported service endpoints at

https

//cloud.google.com/vertex-ai/docs/reference/rest#rest_endpoints

job_resource_name: str¶

The name of the batch prediction job resource, in a form of projects/{project}/locations/{location}/batchPredictionJobs/{batchPredictionJob}. For more details, see

https

//cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.batchPredictionJobs/get

bigquery_output_table: str | None = None¶

The name of the BigQuery table created, in predictions_<timestamp> format, into which the prediction output is written. For more details, see

https

//cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.batchPredictionJobs#outputinfo

bigquery_output_dataset: str | None = None¶

The path of the BigQuery dataset created, in

bq

//projectId.bqDatasetId format, into which the prediction output is written. For more details, see

https

//cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.batchPredictionJobs#outputinfo

gcs_output_directory: str | None = None¶

The full path of the Cloud Storage directory created, into which the prediction output is written. For more details, see

https

//cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.batchPredictionJobs#outputinfo

schema = 'title: google.VertexBatchPredictionJob\ntype: object\nproperties:\n resourceName:\n type: string\n bigqueryOutputTable:\n type: string\n gcsOutputDirectory:\n type: string\n bigqueryOutputDataset:\n type: string'¶

schema_title = 'google.VertexBatchPredictionJob'¶

schema_version = '0.0.1'¶

class google_cloud_pipeline_components.types.artifact_types.VertexDataset(name: str | None = None, uri: str | None = None, metadata: dict | None = None)[source]¶

Bases: Artifact

An artifact representing a Vertex AI Dataset resource.

Methods:

create(name, uri, dataset_resource_name)

Create a VertexDataset artifact instance.

Attributes:

schema
schema_title
schema_version

classmethod create(name: str, uri: str, dataset_resource_name: str) → VertexDataset[source]¶

Create a VertexDataset artifact instance.

Parameters¶

name: str¶

The artifact name.

uri: str¶

the Vertex Dataset resource uri, in a form of

https

//{service-endpoint}/v1/projects/{project}/locations/{location}/datasets/{datasets_name}, where {service-endpoint} is one of the supported service endpoints at

https

//cloud.google.com/vertex-ai/docs/reference/rest#rest_endpoints

dataset_resource_name: str¶

The name of the Dataset resource, in a form of projects/{project}/locations/{location}/datasets/{datasets_name}. For more details, see

https

//cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.datasets/get

schema = 'title: google.VertexDataset\ntype: object\nproperties:\n resourceName:\n type: string'¶

schema_title = 'google.VertexDataset'¶

schema_version = '0.0.1'¶

class google_cloud_pipeline_components.types.artifact_types.VertexEndpoint(name: str | None = None, uri: str | None = None, metadata: dict | None = None)[source]¶

Bases: Artifact

An artifact representing a Vertex AI Endpoint resource.

Methods:

create(name, uri, endpoint_resource_name)

Create a VertexEndpoint artifact instance.

Attributes:

schema
schema_title
schema_version

classmethod create(name: str, uri: str, endpoint_resource_name: str) → VertexEndpoint[source]¶

Create a VertexEndpoint artifact instance.

Parameters¶

name: str¶

The artifact name.

uri: str¶

the Vertex Endpoint resource uri, in a form of

https

//{service-endpoint}/v1/projects/{project}/locations/{location}/endpoints/{endpoint}, where {service-endpoint} is one of the supported service endpoints at

https

//cloud.google.com/vertex-ai/docs/reference/rest#rest_endpoints

endpoint_resource_name: str¶

The name of the Endpoint resource, in a form of projects/{project}/locations/{location}/endpoints/{endpoint}. For more details, see

https

//cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.endpoints/get

schema = 'title: google.VertexEndpoint\ntype: object\nproperties:\n resourceName:\n type: string'¶

schema_title = 'google.VertexEndpoint'¶

schema_version = '0.0.1'¶

class google_cloud_pipeline_components.types.artifact_types.VertexModel(name: str | None = None, uri: str | None = None, metadata: dict | None = None)[source]¶

Bases: Artifact

An artifact representing a Vertex AI Model resource.

Methods:

create(name, uri, model_resource_name)

Create a VertexModel artifact instance.

Attributes:

schema
schema_title
schema_version

classmethod create(name: str, uri: str, model_resource_name: str) → VertexModel[source]¶

Create a VertexModel artifact instance.

Parameters¶

name: str¶

The artifact name.

uri: str¶

the Vertex Model resource uri, in a form of

https

//{service-endpoint}/v1/projects/{project}/locations/{location}/models/{model}, where {service-endpoint} is one of the supported service endpoints at

https

//cloud.google.com/vertex-ai/docs/reference/rest#rest_endpoints

model_resource_name: str¶

The name of the Model resource, in a form of projects/{project}/locations/{location}/models/{model}. For more details, see

https

//cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.models/get

schema = 'title: google.VertexModel\ntype: object\nproperties:\n resourceName:\n type: string'¶

schema_title = 'google.VertexModel'¶

schema_version = '0.0.1'¶