| December 2024 |
20.00.00.03 |
- teradataml no longer supports setting auth_token using set_config_params(). Use set_auth_token() to set the token.
- set_auth_token() - Added base_url parameter which accepts the CCP url.
- FeatureStore support added to handle feature management within the environment
- Added support for lightGBM package through OpensourceML (OpenML) feature.
- New functions:
- register()
- call_udf()
- list_udfs()
- deregister()
- DataFrame updates:
- New alias() function
- New db_object_name property
- Updated join() to support compound ColumExpression having more than one binary operator and/or ColumExpression containing FunctionExpression(s) in on argument. self-join now expects aliased DataFrame in other argument.
- GeoDataFrame updates:
- New alias() function
- Updated join() to support compound ColumExpression having more than one binary operator and/or ColumExpression containing FunctionExpression(s) in on argument. self-join now expects aliased DataFrame in other argument.
- New DataFrameColumn a.k.a. ColumnExpression functions:
- Arithmetic Functions - DataFrameColumn.isnan(), DataFrameColumn.isinf(), DataFrameColumn.isfinite()
- New Table Operator function - Image2Matrix()
- New File Management function - list_files()
- Configuration Option update: table_operator specifies the name of table operator.
- AutoML update - AutoML, AutoRegressor, and AutoClassifier now support DECIMAL datatype as input.
|
| October 2024 |
20.00.00.02 |
- New functions
- udf()
- materialize()
- create_temp_view()
- DataFrame updates:
- New functions:
- set_session_param()
- unset_session_param()
- Updates:
- DataFrameColumn.cast(): Accepts 2 new arguments format and timezone.
- DataFrame.assign(): Accepts ColumnExpressions returned by udf().
- to_pandas() - Returns the pandas dataframe with Decimal column types as float instead of object.
- New teradataml DataFrameColumn a.k.a. ColumnExpression functions:
- _Date Time Functions_
- DataFrameColumn.to_timestamp()
- DataFrameColumn.extract()
- DataFrameColumn.to_interval()
- _String Functions_
- DataFrameColumn.parse_url()
- _Arithmetic Functions _
- AutoML updates:
- evaluate() method added to AutoML(), AutoRegressor(), and AutoClassifier():
- New functions added: load(), deploy(), remove_saved_model(), and model_hyperparameters()
- AutoML(), AutoRegressor(): New performance metrics added for task type regression i.e., "MAPE", "MPE", "ME", "EV", "MPD", and "MGD".
- AutoML(), AutoRegressor(), and AutoClassifier():
- New arguments added: volatile, persist.
- predict() - Data input is now mandatory for generating predictions. Default model evaluation has been removed.
- set_config_params(): ues_url and auth_token arguments will be deprecated in a future release
- list_td_reserved_keywords() accepts a list of strings as argument.
|
| August 2024 |
20.00.00.01 |
- Streamlined the installation process to include wheel file installation option
- Added max_models argument to AutoML
- Added new AutoML example
|
| March 2024 |
20.00.00.00 |
- Add new section of AutoML in teradataml
- Added new section of teradataml Open-Source Machine Learning Functions
- Added new deploy() method to both Script table operator and Apply table operator
- Updated DataFrame Manipulation functions:
- Added new functions cube, rollup, replace
- Updated existing functions join
- Updated fit methods for both GridSearch and RandomSearch, and added new examples for these two algorithms in Hyperparameter Tunning section
- Added new DataFrame Column functions categories:
- Regular Arithmetic Functions
- String Functions
- Bit-Byte Manipulation Functions
- Regular Expression Functions
- Date Time Functions
- Comparison Functions
- Trigonometric Functions
- Hyperbolic Functions
- Updated User Permission in Vantage
- Updated dependencies requirements
- Removed Model Cataloging section
|