United Arab Emirates.

Argentina.

ARIMA model.

New name for the ARIMA model.

Austria.

Australia.

Automatically inferred from timestamps.

Splits data automatically: Uses NO_SPLIT if the data size is small.

[Beta] AutoML Tables classification model.

[Beta] AutoML Tables regression model.

Uses an iterative batch gradient descent algorithm.

Belgium.

Boosted tree classifier model.

Boosted tree regressor model.

Brazil.

Canada.

Switzerland.

Chile.

China.

Colombia.

Use a constant learning rate.

Cosine distance.

Czechoslovakia.

Splits data with the user provided tags.

Czech Republic.

Daily data.

Germany.

Denmark.

DNN classifier model.

DNN regressor model.

Algeria.

Ecuador.

Estonia.

Egypt.

Europe, the Middle East and Africa.

Spain.

Eculidean distance.

Use nonweighted-als for explicit feedback problems.

Finland.

France.

Great Britain (United Kingdom).

Global.

Greece.

Hong Kong.

Holiday region unspecified.

Hourly data.

Hungary.

Indonesia.

Ireland.

Israel.

Use weighted-als for implicit feedback problems.

India.

Iran.

Italy.

Japan and Asia Pacific: Korea, Greater China, India, Australia, and New Zealand.

Japan.

K-means clustering model.

Initializes the centroids using data specified in kmeans_initialization_column.

Unspecified initialization method.

Initializes with kmeans++.

Initializes the centroids randomly.

Korea (South).

Latin America and the Caribbean.

Use line search to determine learning rate.

Linear regression model.

Logistic regression based classification model.

Latvia.

Morocco.

Matrix factorization model.

Mean log loss, used for logistic regression.

Mean squared loss, used for linear regression.

Monthly data.

Mexico.

Malaysia.

North America.

Nigeria.

Netherlands.

Norway.

Data split will be skipped.

Uses a normal equation to solve linear regression problem.

New Zealand.

Peru.

Per-minute data.

Philippines.

Pakistan.

Poland.

Portugal.

Quarterly data.

Splits data randomly.

Romania.

Serbia.

Russian Federation.

Saudi Arabia.

Sweden.

Daily period, 24 hours.

Monthly period, 30 days or irregular.

No seasonality.

Quarterly period, 90 days or irregular.

Weekly period, 7 days.

Yearly period, 365 days or irregular.

Splits data sequentially.

Singapore.

Slovenia.

Slovakia.

An imported TensorFlow model.

Thailand.

Turkey.

Taiwan.

Ukraine.

United States.

Venezuela.

Viet Nam.

Weekly data.

Yearly data.

South Africa.

Encoded as a list with types matching Type.array_type.

Encoded as a decimal string.

Encoded as a boolean "false" or "true".

Encoded as a base64 string per RFC 4648, section 4.

Encoded as RFC 3339 full-date format string: 1985-04-12.

Encoded as RFC 3339 full-date "T" partial-time: 1985-04-12T23:20:50.52.

Encoded as a number, or string "NaN", "Infinity" or "-Infinity".

Encoded as WKT.

Encoded as a string in decimal format.

Encoded as fully qualified 3 part: 0-5 15 2:30:45.6.

Encoded as a string.

Encoded as a decimal string.

Encoded as a string value.

Encoded as a list with fields of type Type.struct_type[i].

Encoded as RFC 3339 partial-time format string: 23:20:50.52.

Encoded as an RFC 3339 timestamp with mandatory "Z" time zone string: 1985-04-12T23:20:50.52Z.

Invalid type.

Enum value maps for Model_DataFrequency.

Enum value maps for Model_DataFrequency.

Enum value maps for Model_DataSplitMethod.

Enum value maps for Model_DataSplitMethod.

Enum value maps for Model_DistanceType.

Enum value maps for Model_DistanceType.

Enum value maps for Model_FeedbackType.

Enum value maps for Model_FeedbackType.

Enum value maps for Model_HolidayRegion.

Enum value maps for Model_HolidayRegion.

Enum value maps for Model_KmeansEnums_KmeansInitializationMethod.

Enum value maps for Model_KmeansEnums_KmeansInitializationMethod.

Enum value maps for Model_LearnRateStrategy.

Enum value maps for Model_LearnRateStrategy.

Enum value maps for Model_LossType.

Enum value maps for Model_LossType.

Enum value maps for Model_ModelType.

Enum value maps for Model_ModelType.

Enum value maps for Model_OptimizationStrategy.

Enum value maps for Model_OptimizationStrategy.

Enum value maps for Model_SeasonalPeriod_SeasonalPeriodType.

Enum value maps for Model_SeasonalPeriod_SeasonalPeriodType.

Enum value maps for StandardSqlDataType_TypeKind.

Enum value maps for StandardSqlDataType_TypeKind.

Aggregate metrics for classification/classifier models.

ARIMA model fitting metrics.

Model evaluation metrics for ARIMA forecasting models.

Model evaluation metrics for a single ARIMA forecasting model.

Arima order, can be used for both non-seasonal and seasonal parts.

Evaluation metrics for binary classification/classifier models.

Confusion matrix for binary classification models.

Evaluation metrics for clustering models.

Message containing the information about one cluster.

Representative value of a single feature within the cluster.

Representative value of a categorical feature.

Represents the count of a single category within the cluster.

Data split result.

Evaluation metrics of a model.

Global explanations containing the top most important features after training.

Explanation for a single feature.

Evaluation metrics for multi-class classification/classifier models.

Confusion matrix for multi-class classification models.

A single entry in the confusion matrix.

A single row in the confusion matrix.

Evaluation metrics used by weighted-ALS models specified by feedback_type=implicit.

Evaluation metrics for regression and explicit feedback type matrix factorization models.

Information about a single training query run for the model.

Information about a single iteration of the training run.

(Auto-)arima fitting result.

Arima coefficients.

Arima model information.

Information about a single cluster for clustering model.

Options used in model training.

Id path of a model.

The type of a variable, e.g., a function argument.

A field or a column.

A table type.

UnimplementedModelServiceServer can be embedded to have forward compatible implementations.

ModelServiceClient is the client API for ModelService service.

ModelServiceServer is the server API for ModelService service.

Type of supported data frequency for time series forecasting models.

Indicates the method to split input data into multiple tables.

Distance metric used to compute the distance between two points.

Indicates the training algorithm to use for matrix factorization models.

Type of supported holiday regions for time series forecasting models.

Indicates the method used to initialize the centroids for KMeans clustering algorithm.

Indicates the learning rate optimization strategy to use.

Loss metric to evaluate model training performance.

Indicates the type of the Model.

Indicates the optimization strategy used for training.