4.4 Supported Data Types for AutoML

AutoML supports the following data types.

Numeric Data Types

• DECIMAL

• DOUBLE

• FLOAT

• INT

• INT UNSIGNED

• TINYINT

• TINYINT UNSIGNED

• SMALLINT

• SMALLINT UNSIGNED

• MEDIUMINT

• MEDIUMINT UNSIGNED

• BIGINT

• BIGINT UNSIGNED

Temporal Data Types

• DATE

• TIME

• DATETIME

• TIMESTAMP

• YEAR

String and Text Data Types

• VARCHAR

• CHAR

• TINYTEXT

• TEXT

• MEDIUMTEXT

• LONGTEXT

Data Type Limitations

AutoML uses TfidfVectorizer to pre-process TINYTEXT, TEXT, MEDIUMTEXT, and LONGTEXT, and appends the results to the data set. AutoML has the following limitations for text usage:

• The ML_PREDICT_TABLE ml_results column contains the prediction results and the data. This combination must be fewer than 65,532 characters.

• AutoML only supports datasets in the English language.

• AutoML does not support text columns with NULL values.

• AutoML does not support a text target column.

• AutoML does not support recommendation tasks with a text column.

• For the forecasting task, endogenous_variables cannot be text.

What's Next

• Learn more about the following:

  • Additional AutoML Requirements

  • AutoML Privileges

• Learn how to Create a Machine Learning Model.

• Review Machine Learning Use Cases to create machine learning models with sample datasets.