The CREATE MODEL statement for transform-only models

This document describes the CREATE MODEL statement for creating transform-only models in BigQuery. Transform-only models use the TRANSFORM clause to apply preprocessing functions to input data and return the preprocessed data. Transform-only models decouple data preprocessing from model training, making it easier for you to capture and reuse a set of data preprocessing rules.

You can use a transform-only model in conjunction with the ML.TRANSFORM function to provide preprocessed data to other models:

  • You can use it in the query statement when creating another model, in order to use the transformed data as the training data for that model.
  • You can use it in the query statement argument of the ML.PREDICT function, in order to provide data for prediction that is processed in the way the target model expects.

For batch feature transformations, it is better to use transform-only models because it lets you process large amounts of data in a short time. For online feature transformations, it is better to use Vertex AI Feature Store because it provides responses with low latency.

You can also use a transform-only model with the ML.FEATURE_INFO function in order to return information about feature transformations in the model.

For more information about supported SQL statements and functions for this model, see End-to-end user journeys for ML models.

CREATE MODEL syntax

{CREATE MODEL | CREATE MODEL IF NOT EXISTS | CREATE OR REPLACE MODEL} model_name
TRANSFORM (select_list)
OPTIONS(MODEL_TYPE = 'TRANSFORM_ONLY')
AS query_statement

CREATE MODEL

Creates and trains a new model in the specified dataset. If the model name exists, CREATE MODEL returns an error.

CREATE MODEL IF NOT EXISTS

Creates and trains a new model only if the model doesn't exist in the specified dataset.

CREATE OR REPLACE MODEL

Creates and trains a model and replaces an existing model with the same name in the specified dataset.

model_name

The name of the model you're creating or replacing. The model name must be unique in the dataset: no other model or table can have the same name. The model name must follow the same naming rules as a BigQuery table. A model name can:

  • Contain up to 1,024 characters
  • Contain letters (upper or lower case), numbers, and underscores

model_name is not case-sensitive.

If you don't have a default project configured, then you must prepend the project ID to the model name in the following format, including backticks:

`[PROJECT_ID].[DATASET].[MODEL]`

For example, `myproject.mydataset.mymodel`.

select_list

You can pass columns from query_statement through to model training without transformation by either using *, * EXCEPT(), or by listing the column names directly.

Not all columns from query_statement are required to appear in the TRANSFORM clause, so you can drop columns appearing in query_statement by omitting them from the TRANSFORM clause.

You can transform inputs from query_statement by using expressions in select_list. select_list is similar to a normal SELECT statement. select_list supports the following syntax:

  • *
  • * EXCEPT()
  • * REPLACE()
  • expression
  • expression.*

The following cannot appear inside select_list:

  • Aggregation functions.
  • Non-BigQuery ML analytic functions. For more information about supported functions, see Manual feature preprocessing.
  • UDFs.
  • Subqueries.
  • Anonymous columns. For example, a + b as c is allowed, while a + b isn't.

The output columns of select_list can be of any BigQuery supported data type.

If present, the following columns must appear in select_list without transformation:

  • label
  • data_split_col
  • kmeans_init_col
  • instance_weight_col

If these columns are returned by query_statement, you must reference them in select_list by column name outside of any expression, or by using *. You can't use aliases with these columns.

MODEL_TYPE

Syntax

MODEL_TYPE = 'TRANSFORM_ONLY'

Description

Specify the model type. This option is required.

query_statement

The GoogleSQL query that contains the data to preprocess. The statistics that are calculated when transforming this data are applied to the input data of any functions that you use the model with.

Examples

The following examples show how to create and use transform-only models.

Example 1

The following example creates a model named transform_model in mydataset in your default project. The model transforms several columns from the BigQuery public table bigquery-public-data.ml_datasets.penguins:

CREATE MODEL `mydataset.transform_model`
  TRANSFORM(
    species,
    island,
    ML.MAX_ABS_SCALER(culmen_length_mm) OVER () AS culmen_length_mm,
    ML.MAX_ABS_SCALER(culmen_depth_mm) OVER () AS culmen_depth_mm,
    ML.MAX_ABS_SCALER(flipper_length_mm) OVER () AS flipper_length_mm,
    sex,
    body_mass_g)
  OPTIONS (
    model_type = 'transform_only')
AS (
  SELECT *
  FROM `bigquery-public-data.ml_datasets.penguins`
);

Example 2

The following example creates a model named mymodel in mydataset in your default project. The model is trained on data that is preprocessed by using a transform-only model:

CREATE MODEL `mydataset.mymodel`
OPTIONS
  ( MODEL_TYPE = 'LINEAR_REG',
    MAX_ITERATIONS = 5,
    INPUT_LABEL_COLS = ['body_mass_g'] ) AS
SELECT
  *
FROM
  ML.TRANSFORM(
    MODEL `mydataset.transform_model`,
    TABLE `bigquery-public-data.ml_datasets.penguins`)
WHERE body_mass_g IS NOT NULL;