Teradata Vantage is our flagship analytic platform offering, which evolved from our industry-leading Teradata Database. Until references in content are updated to reflect this change, the term Teradata Database is synonymous with Teradata Vantage.
Teradata Advanced SQL Engine (was NewSQL) is a core of Teradata Vantage, based on our best-in-class Teradata Database processing capability. Advanced SQL refers to the ability to run advanced analytic functions beyond that of standard SQL.
This document provides detailed description and complete usage information for all the functions and verbs in the Teradata® R Package, tdplyr.
tdplyr Function Categories
Categories of tdplyr package functions:
- td_<name>_<engine>: Analytic functions.
- td_<name>_context: API functions that manage the connection and certain internal data structures called context.
- <name>.Teradata: R S3 functions that override the corresponding dplyr package functions with new functionality. In most cases, tdplyr-specific additional parameters are added to these functions.
- td_<name>_predict_<engine> and predict.td_<name>_<engine>: Scoring and prediction functions that are invoked with the predict S3 generic call or with a corresponding "td_*_predict" function call.
Note: name is part of a function name that indicates the specific task of the function. engine is the location where the function is executed. Valid values are mle for Machine Learning Engine and sqle for SQL Engine.
tdplyr Analytic Function Default Execution Locations
Each analytic function in tdplyr has an equivalent function that can be invoked without engine suffix.
For the tdplyr analytic functions without engine suffix, the following table lists those that have the default execution location in the Advanced SQL Engine.
| Function Name | Executed in | Equivalent to |
|---|---|---|
| td_antiselect | SQL Engine | td_antiselect_sqle |
| td_attribution | SQL Engine | td_attribution_sqle |
| td_decision_forest_predict | SQL Engine | td_decision_forest_predict_sqle |
| td_decision_tree_predict | SQL Engine | td_decision_tree_predict_sqle |
| td_glm_predict | SQL Engine | td_glm_predict_sqle |
| td_naivebayes_predict | SQL Engine | td_naivebayes_predict_sqle |
| td_naivebayes_textclassifier_predict | SQL Engine | td_naivebayes_textclassifier_predict_sqle |
| td_ngramsplitter | SQL Engine | td_ngramsplitter_sqle |
| td_npath | SQL Engine | td_npath_sqle |
| td_pack | SQL Engine | td_pack_sqle |
| td_sessionize | SQL Engine | td_sessionize_sqle |
| td_stringsimilarity | SQL Engine | td_string_similarity_sqle |
| td_svm_sparse_predict | SQL Engine | td_svm_sparse_predict_sqle |
| td_unpack | SQL Engine | td_unpack_sqle |
All other tdplyr analytic functions without engine suffix are executed by default in the Machine Learning Engine, and are equivalent to the td_<name>_mle functions in tdplyr.
Note: Teradata recommends using tdplyr analytic functions with engine suffix.
The Moving Average analytic function does not have an equivalent function that can be invoked without an engine suffix. Invoke on the SQL Engine using td_moving_average_sqle with the argument 'mavgtype'.