td_retrieve_model() | Teradata R Package - 17.00 - td_retrieve_model() - Teradata R Package

Teradata® R Package User Guide

prodname
Teradata R Package
vrm_release
17.00
created_date
November 2020
category
User Guide
featnum
B700-4005-090K

The td_retrieve_model() function allows a user to recreate a tdplyr analytic function object from the information saved with the Model Catalog using td_save_model(). This analytic function object can then be used in the user’s workflow.

For example, if a user has built a user-base classification model, he or she can save the information related to this model and reuse it at every instance when there is need to score a new set of user data.

The td_retrieve_model() function can be used to retrieve only those models that are accessible to the requesting user.

See td_publish_model() for information on granting access to a model to other users.

Required arguments:

name specifies the name of the model to be retrieved.

Example Prerequisites

Follow the steps in td_save_model() to create and save a model.

Example

  • Retrieve the saved model.
    > retrieved_glm_model <- td_retrieve_model("glm_model")
  • Use the retrieved model for scoring.
    > loadExampleData("glmpredict_example", "admissions_test")
    > admissions_test <- tbl(con, "admissions_test")
    > td_glm_predict_out1 <- td_glm_predict_sqle(modeldata = retrieved_glm_model,
                                                 newdata = admissions_test,
                                                 terms = c("id","masters","gpa","stats","programming","admitted"),
                                                 family = "LOGISTIC",
                                                 linkfunction = "LOGIT"
                                                 )
  • Print the scoring results.
    > print(td_glm_predict_out1)
    ############ result ############
     
    # Source:   SQL [?? x 1]
    # Database: [Teradata 16.20.50.01] [Teradata Native Driver 16.20.0.35] [ALICE@Vantage1025/ALICE]
    # Groups: 
          id masters   gpa stats    programming admitted fitted_value
       <int> <chr>   <dbl> <chr>    <chr>          <int>        <dbl>
     1    66 no       3.87 Novice   Beginner           1        0.755
     2    65 yes      3.9  Advanced Advanced           1        0.653
     3    63 no       3.83 Advanced Advanced           1        0.946
     4    61 yes      4    Advanced Advanced           1        0.651
     5    58 no       3.13 Advanced Advanced           1        0.950
     6    56 no       3.82 Advanced Advanced           1        0.946
     7    57 no       3.71 Advanced Advanced           1        0.946
     8    60 no       4    Advanced Novice             1        0.865
     9    64 yes      3.81 Advanced Advanced           1        0.656
    10    68 no       1.87 Advanced Novice             1        0.891
    # ... with more rows