As of January 1, 2020 this library no longer supports Python 2 on the latest released version. Library versions released prior to that date will continue to be available. For more information please visit Python 2 support on Google Cloud.

google.cloud.bigquery.model.Model

class google.cloud.bigquery.model.Model(model_ref: Optional[Union[google.cloud.bigquery.model.ModelReference, str]])

Model represents a machine learning model resource.

See https://cloud.google.com/bigquery/docs/reference/rest/v2/models

Parameters

model_ref – A pointer to a model. If model_ref is a string, it must included a project ID, dataset ID, and model ID, each separated by ..

__init__(model_ref: Optional[Union[google.cloud.bigquery.model.ModelReference, str]])

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__(model_ref)	Initialize self.
from_api_repr(resource)	Factory: construct a model resource given its API representation
to_api_repr()	Construct the API resource representation of this model.

Attributes

best_trial_id	The best trial_id across all training runs.
created	Datetime at which the model was created (None until set from the server).
dataset_id	ID of dataset containing the model.
description	Description of the model (defaults to None).
encryption_configuration	Custom encryption configuration for the model.
etag	ETag for the model resource (None until set from the server).
expires	The datetime when this model expires.
feature_columns	Input feature columns that were used to train this model.
friendly_name	Title of the table (defaults to None).
label_columns	Label columns that were used to train this model.
labels	Labels for the table.
location	The geographic location where the model resides.
model_id	The model ID.
model_type	Type of the model resource.
modified	Datetime at which the model was last modified (None until set from the server).
path	URL path for the model’s APIs.
project	Project bound to the model.
reference	A model reference pointing to this model.
training_runs	Information for all training runs in increasing order of start time.
transform_columns	The input feature columns that were used to train this model.

property best_trial_id: Optional[int]

The best trial_id across all training runs.

Deprecated since version This: property is deprecated!

Read-only.

property created: Optional[datetime.datetime]

Datetime at which the model was created (None until set from the server).

Read-only.

property dataset_id: Optional[str]

ID of dataset containing the model.

property description: Optional[str]

Description of the model (defaults to None).

property encryption_configuration: Optional[google.cloud.bigquery.encryption_configuration.EncryptionConfiguration]

Custom encryption configuration for the model.

Custom encryption configuration (e.g., Cloud KMS keys) or None if using default encryption.

See protecting data with Cloud KMS keys in the BigQuery documentation.

property etag: Optional[str]

ETag for the model resource (None until set from the server).

Read-only.

property expires: Optional[datetime.datetime]

The datetime when this model expires.

If not present, the model will persist indefinitely. Expired models will be deleted and their storage reclaimed.

property feature_columns: Sequence[google.cloud.bigquery.standard_sql.StandardSqlField]

Input feature columns that were used to train this model.

Read-only.

property friendly_name: Optional[str]

Title of the table (defaults to None).

classmethod from_api_repr(resource: Dict[str, Any]) → google.cloud.bigquery.model.Model

Factory: construct a model resource given its API representation

Parameters

resource – Model resource representation from the API

Returns

Model parsed from resource.

property label_columns: Sequence[google.cloud.bigquery.standard_sql.StandardSqlField]

Label columns that were used to train this model.

The output of the model will have a predicted_ prefix to these columns.

Read-only.

property labels: Dict[str, str]

Labels for the table.

This method always returns a dict. To change a model’s labels, modify the dict, then call Client.update_model. To delete a label, set its value to None before updating.

property location: Optional[str]

The geographic location where the model resides.

This value is inherited from the dataset.

Read-only.

property model_id: Optional[str]

The model ID.

property model_type: str

Type of the model resource.

Read-only.

property modified: Optional[datetime.datetime]

Datetime at which the model was last modified (None until set from the server).

Read-only.

property path: Optional[str]

URL path for the model’s APIs.

property project: Optional[str]

Project bound to the model.

property reference: Optional[google.cloud.bigquery.model.ModelReference]

A model reference pointing to this model.

Read-only.

to_api_repr() → Dict[str, Any]

Construct the API resource representation of this model.

Returns

Model reference represented as an API resource

property training_runs: Sequence[Dict[str, Any]]

Information for all training runs in increasing order of start time.

Dictionaries are in REST API format. See: https://cloud.google.com/bigquery/docs/reference/rest/v2/models#trainingrun

Read-only.

property transform_columns: Sequence[google.cloud.bigquery.model.TransformColumn]

The input feature columns that were used to train this model. The output transform columns used to train this model.

See REST API: https://cloud.google.com/bigquery/docs/reference/rest/v2/models#transformcolumn

Read-only.